When considering a new fiber to the home (FTTH) project, people often ask which technology they should use. My initial response is usually a technically savvy, "It depends." A number of technical FTTH network alternatives
The choices generally fall into two categories: passive or active.
Passive optical networks (PONs) have no powered components between the end subscriber and the main distribution point such as a central office or headend. The key characteristic of PONs is that their fiber networks utilize a point-to-multipoint architecture that looks like a tree structure. Passive optical splitters are used to divide the signal among multiple subscribers, typically 32 or 64. The fiber between the last splitter and the end subscriber is dedicated to the subscriber, but all other fiber is shared.
The most common types of PONs are Gigabit PON (GPON) and Ethernet PON (EPON). The current ITU-T G.984 GPON standard has an asymmetrical bandwidth capacity of 2.488 Gbps downstream and 1.244 Gbps upstream. The IEEE 802.3 EPON standard has symmetrical bandwidth capacity of 1 Gbps upstream and downstream. The standards for both PON technologies, however, are now being updated to support 10 Gbps speeds.
Active optical networks (AONs), also called Active Ethernet or Active-E in the world of FTTH networks, utilize traditional Ethernet technologies in a point-to-point star topology. One or more fibers are dedicated to each end subscriber between the subscriber's premise and the first tier of active Ethernet switching/routing equipment. This equipment may be distributed and powered out in the field, or it may be colocated at the main distribution point. With AONs, any standard Ethernet fiber optic transceiver may be used now for capacity up to 1 Gbps or even 10 Gbps and eventually up to 40 Gbps and beyond.
Fiber to the home deployment technology and cost considerations
If cost is a primary consideration for a new FTTH network project, PONs should be carefully considered. Historically, PONs have been more prevalent in the U.S., while AONs have been more popular in Europe and Asia.
If the FTTH project will be utilizing IP services such as IPTV and voice-over-IP (VoIP) along with standard Internet data services, either PON or AON technologies can be used effectively.
President, Knowledge Works LLC
Because PONs usually require less fiber than AONs, and the average distance between subscribers in the U.S. is larger, PONs are the more cost-effective choice for financing FTTH deployments. PONs tend to be the least expensive option overall, no matter what the location, especially for larger carriers serving primarily residential markets that have reasonable take rates.
GPON may be the only viable choice if the FTTH project will be deploying analog/digital RF cable TV or traditional voice services (POTS). GPON is the only cost-effective option that allows these services to be distributed over the same fiber as Internet data to the end subscriber. EPONs and AONs can support IP-based services.
If the FTTH project will be utilizing IP services such as IPTV and voice over IP (VoIP) along with standard Internet data services, then either PON or AON technologies can be used effectively. Other factors would need to be considered in order to make the best choice.
AONs, for example, are gaining in popularity in the U.S. because they can provide increased service flexibility, including better support for media-rich applications requiring high-volume symmetrical bandwidth. Despite their higher relative costs, AONs can be an excellent option for smaller independent operators that are deploying an all-IP suite of services for voice, data and video, and targeting a mix of customers that may include commercial buildings, academic/government campuses, residential multi-dwelling units (MDUs) and higher-density single-family developments.
Future-proofing fiber to the home deployments
Operations and maintenance lifecycle costs also should be considered in any FTTH technology decision, along with the expected costs for upgrading to newer technologies like 10 Gbps. The majority of the costs in an FTTH deployment are associated with the outside plant (material costs for conduit and fiber and installation costs). With a life-expectancy of 25 to 40 years or more, it is important to protect this investment and future-proof it as much as possible.
One future-proofing technique is to build the fiber infrastructure in such a way that it is technology-agnostic as future needs and technical capabilities change. Some projects are being designed to better future-proof their fiber infrastructure by implementing a point-to-point star topology even for PON deployments (by moving the splitters back to the main distribution point). This design enables the fiber infrastructure to be technology agnostic as future needs and technical capabilities change, and even provides the ability to switch from a PON to an AON implementation or vice versa just by changing out the optical/electronic equipment and leaving the fiber untouched.
Another future-proofing technique is to increase the likelihood that the fiber infrastructure itself can support higher data rates in the future by designing and installing it with the lowest practical optical loss. To achieve this goal, techniques such as fusion splicing can be employed along with such materials as low-loss connectors and bend-insensitive fiber.
Making the right technology choice for a new FTTH network project can be a daunting task, but with careful consideration and a clear objective, you can make a good decision. The good news is that there are really no bad choices. All of the choices are proven technologies that can support a wide array of services. It is just a matter of choosing the solution that best meets the needs of the project.
About the author: David Hashman, president of Knowledge Works LLC, is a networking and IT professional who has focused on the fiber-to-the-home market in the U.S. and Mexico for the past six years. Knowledge Works provides management and technology consulting and telecommunications design, engineering and installation services to corporate clients. Current projects include the triple-play design, engineering and installation of a private Gigabit Passive Optical Network (GPON) solution and a private active Ethernet network with a Digital RF Video Overlay. Hashman teaches network and computer courses at a variety of colleges and universities in Colorado. He earned a master's degree in Engineering-Economic Systems from Stanford University and a master's degree in Business Administration from the University of Northern Colorado. He can be reached at firstname.lastname@example.org.
This was first published in September 2010