Estimated reading time - 10 min
- Second generation G.Fast 212 technology – which brings 2 Gbps throughput to legacy copper wire infrastructures — is expected to mature rapidly as telco network service providers (NSPs) look for cost-effective upgrade solutions in established “brownfield” communities.
- 10G-PON technologies are emerging as the future-proof infrastructure of choice for new “greenfield” communities that are being built to address the explosion of data and complexity that will define the connected home in the years to come.
- Technicolor, known as Vantiva, is developing customer premises equipment (CPE) offerings that are on the forefront of both technologies, offering NSPs around the world an array of options as they design, implement and manage their network modernization strategies.
Few things are as certain in life as the ever-increasing demand for bandwidth in today’s global digital society. This helps explain the rapid pace of development in technologies that are breathing life into legacy infrastructures, as new ground is broken to deploy next-generation fiber-to-the-home solutions.
We caught up with Karel Adriaensen, Director of Product Management with Technicolor’s Connected Home Division, to get his insights into the latest developments on two compelling broadband access technologies that are maturing rapidly as telcos around the world wrestle with how to bring more bandwidth in to people’s home: G.Fast and 10G-PON.
Karel, let’s start with understanding the recent progress that is taking place with G.Fast technologies to enhance the capabilities of today’s copper wire infrastructure?
Adriaensen: G.Fast is going through interesting evolutions since we took this technology through its first prototype only a few years ago. Since then we have gone through many trials, and the technology has really matured and is now ready for mass deployment. Starting in 2018, we expect to see the first commercial deployments of first generation G.Fast technology, bringing gigabit performance to this decade’s old legacy infrastructure.
At the same time, a second generation of this G.Fast technology – G.Fast 212 – is being developed that makes use of the extended 212 Megahertz spectrum. This will allow network service providers (NSPs) to double the throughput that can be achieved over the conventional copper wire to 2 Gbps.
This creates quite a lot of new possibilities for NSPs to think about as they assess how to use this technology as they develop their long-term broadband deployment plans.
For one thing, G.Fast breathes new life into the existing copper wire infrastructure that has been in place — in some cases — for decades.
To be clear, the G.Fast technology is able to achieve these rates over fairly short distances. So, for instance, where ADSL was able to maintain its level of performance – of up to 24 megabits per second (Mbps)] for over 5 kilometers – the second generation of G.Fast can support the 2 Gbps performance for 50 to 100 meters maximum.
The idea is to push the fiber infrastructure as close as possible to consumers’ homes, and then make the transition to the infrastructure that is already touching residences to complete the connection at gigabit-to-multi-gigabit speeds over conventional copper.
Meanwhile, on the fiber optic side of the equation, we are seeing dramatic progress in improving the performance of passive optical networking (PON) technologies with the introduction of fiber solutions that can deliver 10 Gbps capabilities. Can you tell us a little bit about what is happening with the new generation of 10G-PON technology?
Adriaensen:That is a very interesting technology as well. 10 G-PON is gaining a lot of attention from telco NSPs in North America.
Telcos have observed with interest how cable NSPs have adopted Data Over Cable Service Interface Specification (DOCSIS) technologies, first with DOCSIS 3.0 and most recently with DOCSIS 3.1 – which allows cable providers to deliver performance that — in some cases – exceeds that 1 Gbps threshold.
So the telcos had to react and respond. Telco NSPs have done so by developing 10G-PON technology. Moving comprehensively to this technology will, of course, require several years to achieve, but the process is already in motion. For instance, we are seeing quite a bit of activity already in North America.
Other parts of the world are also taking an interest. In Europe, as an example, more and more NSPs are thinking about this 10G-PON option.
So under what circumstances are telco NSPs looking at technologies like 10G-PON, and when are the looking at the latest generation of G.Fast technologies?
Adriaensen: That is an interesting question. So, the real sweet spot for 10G-PON is when there is a “greenfield” opportunity – such as when a new community is being built, and there is no pre-existing infrastructure in place that can bring connectivity to these new homes. In this scenario, since NSPs have to provide a connection from scratch anyway, they might as well bring fiber to the home right from the start.
10G-PON is a very attractive solution that has the added advantage of being future-proof for decades to come.
The situation is different in a “brownfield” situation, such as when you are in the middle of an established community. In this case offering a fiber-based gigabit or multi-gigabit capability is a much more cumbersome proposition.
And the most cumbersome part of upgrading services in this environment is the last few meters as you approach the home. This is when you must start digging through gardens, secure municipal licenses and get special permissions from landlords so that you can get access to the end users’ home.
In this scenario, G.Fast technology comes in to play very well. The idea here is to bring fiber as close as you can to the end user, and then make the transition to the copper that is already in place by using G.Fast to bring high-speed capabilities into the living unit.
As NSPs develop their infrastructure migration strategies, what do you see as the key issues that they need to manage as they move forward with their plans?
Adriaensen: Well, as you can imagine, anytime you introduce new technologies there are challenges that must be addressed.
With G.Fast — especially the latest generation of this technology that is just now being developed — telcos will be working with the highest frequencies that have ever been used on copper lines. As a result, we as an industry have to go through a learning process — through lab validations and field trials — to understand all the ramifications of this new technology.
We will have to understand, as an example, the impact of the quality of the copper — that is already in place – on data rates. As the second generation of G.Fast matures, we will have to learn how to fine-tune the technology to take variables like this into account. That is a process that we have already gone through with the first generation of G.Fast.
The good news, I believe, is that the second generation of G.Fast will go through this process very quickly. The process will be shorter because the fundamentals of the first and second generation G.Fast technologies are the same. We are simply extending the spectrum that we are using. So, a lot of the things that have been learned in bring the first generation of G.Fast to market can be reused in developing the second generation of this technology. The learning curve will be shorter.
Moving your attention to the 10G-PON technology, what do you suppose are the key issues that the industry will face in bringing this technology to market?
Adriaensen: There are challenges here as well. But they can be readily addressed.
The first thing to keep in mind is that there are more than one flavors of 10G-PON – which is also referred to as XG-PON.
For instance, there is the Next Generation Passive Optical Network 2 (NG-PON2) technology. This is considered one of the most flexible approaches to deploying passive optical networking technologies. It uses four different wavelengths in the fiber which allows NSPs to move to the highest possible data rates.
The problem with NG-PON2 is that it is also one of the most difficult and expensive of the PON technologies to deploy. This is because the ability to support four wavelengths creates extremely complex, high performance – and therefore expensive – requirements for customer premises equipment (CPE).
This is a tough issue for an industry that is always searching for ways to implement the most cost-optimal solution. For this reason, it will probably take quite some years before the costs for supporting this technology have eroded to a level in which deployments for residential applications become economical. That is why NG-PON2, today, is more relevant for enterprise deployments, where higher investments can be justified.
Another high-performance PON variant is known as XGS-PON, which supports 10Gbps speeds, but does so only on only one wavelength. This makes it simpler to implement than NG-PON2. In fact, even though this technology is almost brand new by comparison, we are already seeing strong cost erosions, making it more affordable to deploy for residential applications.
So given these developments, when and where do you believe we will see 10G-PON become a mainstream part of the telco infrastructure for delivering broadband access to the home?
Adriaensen: My view is that we will first see this becoming a mainstream part of telco deployments in North America. A couple of NSPs in this region already have started preparing for a move to this technology over the past two years, and I expect the volume of deployment to pick up significantly in 2018.
In Europe, we are seeing ongoing discussions on this topic pick up, and many NSP in this region are laying out their plans for 10G-PON as well. Over the course of 2019 and 2020, I think you will see significant deployment activity in this region.
In parts of the Middle East, I expect to see activity with this technology in late 2018 and accelerate in 2019.
Then there is Asia. China, today, is the biggest market for GPON technology, and there is a move taking place to adopt 10G-PON. Right now, the focus for 10G-PON in china is on business applications. But as the costs erode, I expect activity to pick up on the residential front, as well, over the next couple of years.
What about G.Fast…which “brownfield” environments do you expect to see the most activity for with this technology?
Adriaensen: The first generation of G.Fast is already seeing a lot of activity. There has been a long learning curve since it was introduced in 2014. It has matured to the extent that next year, in 2018, we will see large scale deployments happening. There are many confirmed plans that have been announced.
The second generation of G.Fast – or G.Fast 212 – is still brand new. But I expect to begin to see broad adoption in late 2019 and for it to pick up speed in 2020 in major urban areas.
So what is Technicolor doing to bring these new broadband access technologies to market?
Adriaensen: Technicolor has been a major player in driving adoption of these technologies by promoting its development through field trials and lab tests.
With regards to G.Fast 212, we have prototypes of CPE that have embedded this technology in our labs. Our engineers have made a lot progress. As a result, we have been able to provide live demonstrations of this technology in our products at the Broadband World Forum 2017 in Berlin.
After that we will be engaged in lab tests and field trials of our G.Fast 212 CPE with our NSP customers.
On the 10G-PON front, we have already been building CPE for this technology for a while now. We are already embedding this technology into our gateway offerings. This technology is currently ready, and we are in the final validation stages with NSPs for commercial deployment in 2018.
We have also integrated our 10G-PON gateway with 802.11ax Wi-Fi capabilities, which is the newest, latest and greatest technology as far as Wi-Fi is concerned. This also has been on display during the BBWF 2017 conference in Berlin.
