Operators around the world are seeking to deploy ultra-high bandwidth solutions throughout their networks to achieve national objectives, global broadband competitiveness and to enable new and exciting applications and services. At the same time, they are also trying to go “green” by selecting products that consume less power. This article describes the solutions providers can employ that move beyond FTTH and FTTN/FTTC to achieve these goals.

The Global Goal: Ubiquitous 100 Mbps Service
Many countries, including the US, UK, France, Germany, Australia, New Zealand, and Finland, have established National Broadband Plans that call for 100 Mbps to a large majority of households by no later than 2020 (some target 2015). These countries have realized that broadband is a key element to the prosperity and health of their nation, and they have set some very aggressive goals to further broadband in their respective nations.

“Broadband is the next tipping point, the next truly transformational technology,” said ITU Secretary-General Dr. Hamadoun Touré. “It can generate jobs, drive growth and productivity, and underpin long-term economic competitiveness. It is also the most powerful tool we have at our disposal in our race to meet the Millennium Development Goals, which are now just 5 years away.”

Broadband speeds of 100 Mbps and beyond make a substantial difference to the user experience -- allowing paradigm shifts and enabling new applications. 100 Mbps throughput allows current applications to work much better -- for example, you can download an entire music album in a handful of seconds or a complete HD movie in only a few minutes. But it can also lead to new services that are simply not possible with today’s broadband: imagine being able to simultaneously watch multiple high-definition camera views of your favorite sporting event in 3-D on your wall of flat-screen displays.

While most of the national broadband plans have concentrated on downstream speeds, fast upstream speeds are also important in enhancing the user experience. In addition to accelerating downloads by speeding the acknowledgements that are part of TCP-IP, a 100 Mbps upstream link (50 to 100 times faster than today’s typical upstream speed) will also contribute to a much better user experience. For example, with a 100 Mbps upstream link, you could upload a high-resolution 4 Megabyte photo in a fraction of a second or a 30-second HD video clip to YouTube in about 10 seconds. Speaking of YouTube, YouTube users are currently uploading 24 hours of video every minute. That might seem like a lot until you look at recent numbers from live streaming services such as Ustream, Livestream and Justin.tv. Each of these encode more hours of video per minute than YouTube, putting substantial pressure on the upstream path. With each of these examples, upstream speed is critical. Users want the ability to upload their content quickly and easily.

But speed isn’t everything. Once you reach speeds of 100 Mbps, latency, the delay in delivering data, is a critical factor for achieving a good user experience in many applications. The combination of high upstream bandwidth and low latency can lead to new revenue-generating services such as residential high-definition telepresence, cloud-based computing, and realistic 3D multiplayer gaming, as well as new services leveraging user-generated content.

While video tends to be top of mind when people think of services delivered over high-speed broadband, gaming is not far behind. The overall worldwide gaming market is now in the tens of billions of dollars and Massively Multi-Player Online Role-Playing Games (MMPORG) represents a major component of that. Such games are very delay-sensitive because people from all over the world are playing together at the same time. In addition, the impact of network delay is about to become a bigger issue with the emergence of cloud gaming. Instead of the game residing on a computer or console, cloud gaming processes everything in the cloud. This means that all HD images and controller actions are transported across the network in real time. It is certainly conceivable that future games will require even more upstream speed and ultra-fast response times.

Imagine, for example, a number of people across the world playing an orchestral version of Rock Band, all complemented by an HD Webcam for each player and with the video feeds merged in the cloud to create the semblance of a live orchestra watched by millions across the world. Sound far-fetched? Maybe, until you consider that YouTube was launched a mere 5 years ago and people now watch two billion videos on YouTube every day.

As we look to 2015 and beyond, it is impossible to guess what the new services will be. However, it is clear the bandwidth requirements will be massive and that user-generated content will play a major role. That means that ultra-fast downstream, ultra-fast upstream, and ultra-low delay will be critical.

Indeed, competition in the broadband space to serve these growing consumer needs will continue to be fierce. Premium services will deliver more revenue for operators and will also allow them to remain highly competitive if they can deliver the bandwidth. Driven by a desire for advanced services and a better overall experience, consumers are constantly demanding more speed and they are voting with their wallets. Competition is heating up for super-fast Internet, with cable operators such as Comcast (US) and Virgin Media (UK) already rolling out 100 Mbps over their Hybrid Fiber/Coax (HFC) network.

Fiber to the Rescue
Wireline operators are fighting back with deep-fiber deployments like Fiber-to-the-Home (FTTH). FTTH technology is a critically important tool for wireline operators, as it can be easily deployed in greenfield and some brownfield scenarios to deliver broadband speeds of 100 Mbps and up. However, FTTH deployments can be prohibitively expensive and time consuming in scenarios where the fiber must be buried in roads, sidewalks and driveways. (See Figure 1.)

Figure 1.

In addition, there are labor and material costs associated with installing the Optical Network Terminal (ONT) unit on the side of each subscriber home and connecting it to the customer’s in-home wiring. For instance, a technician must be scheduled to visit the residence (and have the resident present) to connect the ONT on the side of the house to the home network, turn up the service and test it. The in-home installation often takes at least half a day or more since the effort required to run cable through an existing home can be unpredictable. This is expensive, disruptive to the customer, and ultimately adds another element of delay in getting ultra-fast broadband rolled out across the operator’s network. Use of indoor ONTs may reduce the cost of equipment, but still require an expensive and time-consuming professional installation to extend fiber inside the home.

Operators around the world are deploying FTTH successfully, but they are having a difficult time making the business case work beyond 30 to 50 percent of their networks. For example, Verizon launched the FiOS FTTH initiative in 2004 and now passes close to 50 percent of customers in their 16-state territory. BT started rolling out FTTH in 2010 and 1.5 million homes have been passed so far with a plan to reach 40 percent of homes by 2012. However, in both of these cases, plans to address the remaining homes with fiber have been delayed due to the skyrocketing cost.

FTTH is a great technology but it is clear that a complementary next-generation solution is desperately required to reach the 50 to 70 percent of the network that FTTH cannot reach economically or in a timely fashion. This is especially true in light of the fact that cable operators are currently leveraging high-bandwidth DOCSIS 3.0 technology and their existing Hybrid Fiber/Coax plant to rolling out 100 Mbps services. Time is of the essence.

How About FTTN or FTTC?
The obvious question is: Why not use solutions like FTTN or Fiber-to-the-Cabinet (FTTC) as alternatives to FTTH?

FTTN and FTTC are great broadband solutions which have important roles in most broadband deployments. The DSL technology providers use today can deliver tens of Mbps over thousands of meters, making them ideal for deployment in existing cabinets or cross-connects. These solutions are a valuable part of the service providers’ toolbox. They provide an economical alternative to FTTH in serving areas where getting fiber closer than about a half-mile from the home is prohibitively disruptive and expensive (i.e., buried in backyards near fences, etc.).

By pushing fiber to node (serving typically 48-384 subscribers) and providing DSL (ADSL2 or VDSL2) over the existing twisted pair infrastructure, broadband speeds of up to 50 Mbps downstream are being delivered today, but the upstream speeds are substantially less. By using the emerging technologies of pair bonding and vectoring it should be possible to increase the downstream speeds to 100 Mbps or a bit higher in the future. However, if the copper can be shortened to a few hundred feet, similar technologies can be used to deliver broadband speeds that approach 1 Gbps, which is FTTH territory.

With the proliferation of user-generated content, and high-end multi-player gaming, new services like telepresence and cloud-based services delivering very-fast upstream bandwidth is becoming very important. This could provide operators with a true differentiated service relative to competitors like cable companies.

To deliver ultra-high performance, the electronic equipment needs to be placed very close to end consumers. This implies the need to deploy a very large number of small systems, likely in the order of 8 to 16 ports per system. This presents new challenges.

Challenge 1: Power Delivery.
It's one thing to bring power feeds to tens of cabinets in a particular region. It is quite another to deliver network power to thousands of very small systems.

Challenge 2: Size.
When deploying many small systems close to end consumers, it is critical for the solution to be ultra small so that it easily fits into existing pedestal and footway boxes. It allows the operator to deploy equipment very quickly and cost effectively.

This presents new challenges that FTTN and FTTC were not designed for. When taking ultra-high performance, low power, and small size into consideration, it becomes clear that there is a gap in the operator toolbox: a new product category is required.

This category must deliver ultra-high performance in the order of 100 Mbps or more both upstream and downstream, be packaged in an ultra-small housing, require no network power, be extremely flexible and be compatible with FTTH architectures and products -- all at a fraction of the cost of FTTH. This new product category is called Ultra Broadband Ethernet.

Challenges Turn Into Opportunities
To come with up something radically new, it is sometimes necessary to toss out assumptions and start from scratch, while still leveraging many years of success and experience developing and deploying leading-edge products.

It is also necessary to think differently and ask new questions. The solution started to crystallize when a leading R&D team began looking at this problem from a different perspective: If bringing fiber to the ONT is such a challenge in certain scenarios, could we bring the ONT to the fiber instead? This was the key question which led to the new tactical solution called Ultra Broadband Ethernet.

The team realized that there were locations in the network like Distribution Points (DP) or Subscriber Drop Pedestals that had not been adequately leveraged thus far. This is the location in the network where existing subscriber drops fan out to individual homes. Subscriber pairs are typically easy to access within the DP or pedestal, but space is usually very tight and there is no power. (See Figure 2.)

Figure 2.

The DP is usually only a few hundred feet away from the 8 to 16 homes it serves, and it is often much closer. The team's mission was to develop fiber-fed solutions that would fit in a DP or pedestal and serve at least 8 homes with ultra-fast broadband using the existing subscriber drops. Such a product would allow operators to deploy ultra-fast broadband very quickly and cost effectively.

Furthermore, if it was possible to develop a low-power solution that could be powered from the subscriber premises through the subscriber drop pairs instead of deploying a massive power distribution system, it would save operators even more money, accelerate deployments and help them remain competitive, while also minimizing their carbon footprint.

As indicated earlier, DPs and pedestals tend to be very close to subscriber homes. This is a great location from which to deploy broadband since the loops are so short. However, space within these structures is extremely limited and there is no power available.

In order to rise to the challenges of power and size, innovation on many fronts would be necessary.

Solution 1: Low Power.
The solution to this challenge was to capitalize on very- low power transmission technologies that were developed specifically to deliver very-high bandwidth with low delay over very-short distances: Ethernet. This team's engineers were able to adapt Ethernet transceivers to deliver a minimum of 100 Mbps over existing residential subscriber loops, while allowing the entire product to consume about 1W/line, substantially less than VDSL2-based solutions.

Solution 2: Small Packaging.
The low heat output and the very-high density of the electronics enabled the team to develop a very compact, sealed package. The resulting housing is so small that it can be installed in existing footway boxes and pedestals.

Solution 3: New Powering Techniques.
The last piece of the puzzle was power. It was critical to avoid network power to not only lower costs but also increase deployment speed and flexibility. To solve this huge challenge, an innovative solution was created that delivered power over the subscriber drop pairs from the customer's home to the electronics in the DP.

The same pairs that are used to deliver 100 Mbps or more to the subscriber also double as a power delivery mechanism. Each subscriber would have to deliver on average 1 Watt of power and 4 Watts in the worst case (only one active subscriber) -- equivalent to a child's nightlight.

Now that the team had identified the DP/pedestal as the ideal location for a new product, and had selected a readily available, high-performance, lower-power technology and a way to package it and power it, a viable FTTH alternative was finally emerging.

Ultra Broadband Ethernet consists of a very small fiber-fed multi-port ONT that is installed in a DP or pedestal and serves a small number of subscribers, in the range of 8 to 16. Each subscriber gets a dedicated, point-to-point link that can deliver a minimum of 100 Mbps upstream and 100 Mbps downstream over their existing subscriber drop. (See Figure 3.)

Figure 3. Ultra Broadband Ethernet enables 100 Mbps both upstream and downstream.

Figure 4. Broadband Access Solutions Costs Comparison.

The following elements are included:

Ultra Broadband Ethernet ONT:
The Ultra Broadband Ethernet ONT is packaged in a sealed and submersible housing which can be installed virtually anywhere -- hand holes, footway boxes, pedestals and even telephone poles. It includes a fiber uplink to connect to existing fiber systems.

Media Adapter:
A simple, low-power Media Adapter is used in the home to adapt the external Ethernet signal to an in-home 10/100/1000Base-T which can interface directly to a computer or residential gateway.

The Media Adapter also delivers power to the Ultra Broadband Ethernet ONT at the DP using a low DC voltage over the existing subscriber pair; completely avoiding the need to deliver power from the network. The power required for the multi-port ONT is shared by subscribers, with each subscriber delivering on average 1W of power.

An Enlightened Solution
Competition for ultra-fast broadband is heating up. New services such as cloud gaming, user-generated live streaming, HD 3D TV, multi-player online games, telepresence video conferencing, and others, are pushing the need for high upstream and downstream speeds and ultra-low delays. Cable operators are already rolling out 100 Mbps-based services using DOCSIS 3.0, and wireline operators are fighting back with deep-fiber rollouts using FTTH, FTTN, and FTTC architectures.

FTTH is being successfully used to deliver super-fast broadband and premium services. However, the cost and delays associated with burying fiber, installing the ONT, and turning up the service in a home have limited its deployment to only a portion of the operators' networks. FTTN and FTTC have an important role in most broadband networks, but top out at speeds around 100 Mbps. To help operators deliver broadband speeds of 100 Mbps to 1 Gbps quickly and economically, a new product category is warranted.

Ultra Broadband Ethernet addresses this need, providing operators a new solution for delivering ultra-fast broadband access quickly and economically. Ultra Broadband Ethernet stands on its own as a new product category and a differentiated architecture because of its unique packaging, performance, and powering features.

This new product category allows operators to compete effectively against the onslaught of competitive 100 Mbps services from cable companies, while also helping them reach new customers and deliver high-revenue premium services and bundles. Ultra Broadband Ethernet is real and in trials with several major Tier 1 operators around the world.

Kevin Schneider is Chief Technology Officer for ADTRAN. He leads the ADTRAN corporate technical staff, which is responsible for ADTRAN's research activities, the creation and analysis of new technologies, and participation in industry-wide standards development organizations. He currently serves on the board of the Alliance for Telecommunications Industry Solution (ATIS) where he has chaired the ATIS TOPS Council Optical Access Networks and IPTV Focus Groups, and established ATIS' IPTV Interoperability Forum. For more information, email tammie.dodson@adtran.com or visit www.adtran.com.

What is your experience with this? Tell your fellow readers now!