Pondering PON

Passive optical networking is here to stay--and may be the best hope for the RBOCs to improve their chances against cable.

PON, or passive optical networking technology, has been around for many years—primarily the ATM-based aPON, which was standardized as part of ITU G.983. The frequency of acronyms involved with PON market has increased over the past few months. Now there are terms such as FTTH (fiber to the home), FTTC (fiber to the curb), FTTB (fiber to the business), aPON (ATM-based PON), ePON (Ethernet-based PON), and GPON (Gigabit PONs), plus a general term, FTTP (fiber to the premises). Is this a market of any substance, or is it simply more hype?

PON was designed to provide fiber-based services to homes, apartment buildings (aka multiple dwelling units, or MDUs), and buildings that house a number of small business offices (multiple tenant units, or MTUs). PON is designed to serve these locations that will never have the need for large bandwidth capacities. PON is not designed to serve large buildings or large corporate sites that have large bandwidth needs, where non-PON direct fiber connections are used.

To understand why PON was needed, we take a quick look at the costs of fiber connections, where they are predominantly used today—fiber connects most large buildings. Since there is a lot of business for service providers in each large building, it is worth the considerable expense of $30,000 to $500,000 to lay a mile of bundled fibers. PON is designed for low-cost installation of fiber for smaller buildings with much less revenue potential per building. Large buildings are connected with multiple fibers directly running to the service provider.
 

PON uses a single fiber with many "drops" to reach many buildings, much like a fishing line with multiple hooks used to catch many small fish on a single cast. For large fish, a fisherman uses heavy line and single hook to focus on a single fish. For large buildings, direct fiber runs have an interface (port) on each end. For PON, there is a single port at the provider side, and between eight and 32 ports on the home side, a single port on each drop.

We think this market has promise. Worldwide PON revenue was $23M in 2Q03, up 15% from 1Q03 and we forecast quarterly revenues will double to $46M by 1Q04. By CY06, we project that annual revenue will hit $532M, a compound annual growth rate of 63% from CY01. The main reason for this growth is that service providers want to offer higher bandwidth services on fiber than what can be delivered on copper, believing that eventually nearly all buildings will be connected with fiber.

Service providers around the globe, and particularly in Japan, are beginning to install aPON products and to figure out how to profitably build business and residential services around them. In North America, there will be an increased rollout of FTTP starting soon and taking place over the next 10 to 15 years. Initially, this rollout will take place in new home developments in 2004, and will gradually encompass existing homes and businesses beginning in late 2005. Why?

The RBOCs are interested in fiber now, as they are expecting the FCC to give them unfettered control of fiber access systems. Other reasons RBOCs have jumped on PON are:

• the cable MSO triple threat competition uncomfortably closing in on RBOC customers,

• the fast growing loss of primary line phone service.

In 2Q03, many manufacturers responded to a joint RFI survey from BellSouth, SBC and Verizon, for a major FTTP evaluation. A joint RFP was issued in June 2003, and the RBOCs are deciding on their short list of manufacturers as this column is being written. The RBOCs are targeting lab testing in 3Q03 and initial field
trials in 4Q03 or 1Q04. This joint RBOC process is similar to the joint RFP that started the DSL deployments in 1996. Alcatel won the DSL RFP, and has held a commanding number-one market share position worldwide since.

If the RBOCs deploy fiber to homes en masse, they will improve their competitive position against cable. In particular, the RBOCs will have the access means to offer not only voice and data, but also the final member of the "triple play" triumvirate—video. Fiber to homes is an eventuality, since the inexorably increasing bandwidth trend can eventually only be solved by fiber. The only uncertainty is the timing. aPON will be the PON technology gorilla for the next couple of years, although the arguments for Ethernet suggest eventual ePON or gPON dominance; we forecast aPON to have a significant revenue lead through CY06.

We believe PON is here to stay, in whatever form, today's aPON, or tomorrow's ePON or G-PON. It is an economic fiber solution for service providers to deliver higher bandwidth video, voice and data services to consumers and small businesses that cannot be reasonably delivered on copper.

Michael Howard is Principal Analyst & co-founder of Infonetics Research, Inc., San Jose, CA. He can be reached at 408-298-7999 x223 or at michael@infonetics.com.

Visit Infonetics online at www.infonetics.com.