.gif)
Fibre to the node
Europe presents both opportunities and unique challenges to service providers seeking to build and future-proof robust, flexible FTTN infrastructures, says Joachim Brunzel
Pushing optical fibre closer to end users presents significant new challenges to service providers all over the globe – but none are more unique than those within the European market. Across Europe, infrastructures and methodologies differ from country to country and, to date, most have been quite happy with their physical plant despite virtually no overhead distribution and very little buried fibre cable.
In most cases, there's little likelihood that any new physical plant will be deployed anytime soon. Rather, service providers are seeking the best way to use existing ducted infrastructures to deliver new services and the highest possible speeds to customers. The key for network planners is to consider – in the early stages – what architectures will make the most sense for building a flexible network that will take them into the next decade of high-bandwidth demand.
Ducts unlimited
Much of the existing infrastructure in Europe consists of an intricate system of buried ducts containing copper cabling that connects multiple nodes. Fibre is usually limited to straight cables with single splices that connect two buildings owned by the same service provider. Therefore, a distributed architecture that feeds multiple nodes and offers redundant routes isn't being discussed in too many network planning meetings.
The reason is that building fibre rings where there has never been fibre is an expensive and disruptive operation. Tearing up public and private property is expensive, but there are also challenges to running fibre through existing copper-filled duct lines when the goal is to gain flexibility and easy access. But despite these issues, demand is reaching the point where network planners must make careful decisions about the best method for getting equipment closer to the customer.
Incorporating redundancy into an FTTN network architecture that's deployed through an existing 'tree and branch' duct system presents a huge obstacle. Ignoring the need for redundancy isn't a viable option and risks hefty revenue losses in the event of unexpected outages. Despite these issues, however, most European network planners only have ears for solutions that will enable them to follow traditional tree and branch copper routes.
Why? Because fibre must eventually reach the same locations, ducts already exist to get it there, and it's more cost effective than deploying any new fibre ring. Running a 'thumb-sized' fibre cable in the same ducts that currently accommodate a 'forearm sized' copper cable is not an issue. Therefore, the fibre will have to follow the exact same route – from the CO to the physical cross-connect points in the existing duct network.
In a traditional main cable deployment, a large cable runs from the CO to feed different copper connection points. The fibre deployment for a next-generation network must also hit each copper connection point. Therefore, it must logically follow the same physical path – there are no other viable options.
Typical copper distribution duct systems have multiple ducts leaving the CO and branching in other directions at specified distances. Rather than one cable going from one duct to one group of cabinets, it's multiple ducts with multiple cables sharing the infrastructure for part of the length and then branch off in different directions.
These radial feeds from the CO provide coverage to a nominal circular area of between four and eight cabinets per main cable. The reach from the CO is approximately 4-5 km, dictated by the cable gauge. Two or more main cables might feed in the same direction, varying only in overall length or reach. One main cable would feed the closer cabinets while the other feeds the more distant cabinets. For example, there might be 35,000 copper pairs leaving a CO on 20 main cables of various sizes.
With those 20 cables, providers are able to feed 80 to 100 cabinets.
Migration to fibre
Migrating to an FTTN network requires – as in any geographical area – cutting loops back to less than 5000 feet. ADSL or VDSL services demand that equipment be within 1.5 km of the customer. While the CO can still service customers within that 1.5 km range, two-thirds of a typical 5 km CO service area will require fibre feeds to active cabinets placed closer to the outlying customers. This requires that two-thirds of the cabinets be fibre-fed and active – and it will be achieved by using existing ducts travelling in four different directions before branching out to multiple cabinets.
Since using existing ducts is more cost effective than building new rings, the question of achieving redundancy in a tree and branch architecture still remains. Planners must also decide how many fibre drops per cabinet will ensure enough bandwidth for today as well as tomorrow's passive optical network (PON) upgrades.
That's where good planning comes in. Since fibre counts from cable to cable don't vary much in terms of price points, it's critical to ensure plenty of available fibre. Using smaller feeder cables may also provide some advantages. For instance, winching a 576-count cable through a congested duct is more difficult than pulling a 144-count cable through. Smaller cables may also provide an easier means for achieving redundancy.
Other issues to consider include the age-old consideration of whether to splice or patch (connect) cables. Again, many service providers have their own rules and standards. In a patch, the cable is brought above ground into a patch cabinet. The alternative is to splice it in an underground splice closure.
In Europe, with its typical building-to-building connection, splicing has been the norm. However, providing services to small groups of houses in a tree and branch configuration, the use of patch cabinets and connectors makes more sense in certain locations for achieving maximum flexibility and ease of troubleshooting access.
An all-spliced network could make operational costs soar when technicians must gain access to a particular part of the network. A patch solution where it makes the most sense is the first step in building a more flexible and robust FTTN architecture. Even though existing ducts are being used, planners should create, at a minimum, one main fibre cross-connect (MFCC) at a suitable junction in the physical network.
The MFCC is the most convenient point to bring the fibre above ground to create easier access and improve network flexibility. Again, not every splice or cable should be above the ground – just where it makes sense within the physical infrastructure.
Solving the redundancy issue
Achieving redundancy in a tree and branch network system can be done by first giving consideration to cable size – for example, using two 144-count cables instead of a single 288-count cable. By bringing the two 144-count cables above ground into a fibre cabinet, the tubes in each cable can be split out. By putting six tubes of the first cable onto the second cable, and vice versa, a second functional route is formed downstream. Should a break occur in either feeder cable, a redundant path is available.
Further redundancy can also occur farther downstream. For example, the one main feeder cable passing through the MFCC continues to the SFCC. At this junction, the fibre tubes are split once again to create redundancy from that point downstream to each primary cross-connect point (PCP). Using a 50/50 splitter at each cabinet allows automatic route transfer in the event of a tube or complete cable failure. Since only short distances are involved, loss budget issues are minimal. They are far outweighed by the benefit of achieving a degree of security through redundant cable routes from the CO to the nodes.
Further issues
Additional issues for FTTN network planners to consider include ensuring adequate cabinet space – leaving enough room at each FTTN node for additional splitter modules for future PON upgrades. They may even want to consider using a 90/10 splitter to feed one fibre back to the CO to provide a test field. This would provide technicians the ability to test and monitor every cabinet from a single point. Patch cabinets should also be allowed extra space for future additions, since they could possibly become hubs for future PON configurations.
There is no doubt that bandwidth demand continues to rise, and getting fibre closer to the customer is the only way to meet any future demands once copper capabilities are exhausted. In Europe, capital expense implications are tied to every solution, but they must be weighed against the potential for better operational savings in the future.
Network planners are vital to ensuring the most flexible, robust network architecture at the lowest possible up-front expense. The decisions they make today will greatly affect the ability of their FTTN network to meet every future demand for the next generation of high-bandwidth services.
• Joachim Brunzel is the product manager for ADC Krone's fiber-optic/carrier division joachim.brunzel@adckrone.com
Share this article with others
Post to del.icio.us
Printed from http://www.eurocomms.com/features/111211/Fibre_to_the_node.html
Comment on this article
Skip to comments
We encourage users to analyse, comment on and even challenge European Communications's articles, including the one above - 'Fibre to the node'
User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site.
Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. We will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site.