By C. David Chaffee, Founder and CEO, Chaffee Fiber Optics
For Vincent O’Byrne, director of wireline access technologies at Verizon, OFCNFOEC 2012 is a place where he can check on the status of fiber-to-the-premise deployments around the world. “I am definitely interested in seeing how everyone else deploys FTTP,” says Vincent. “It is always interesting and it may help us to save on our cost structure going forward.” You will probably find Vincent at the session on Monday, March 5 on Access System Trials, where there will be speakers from Germany, Japan, Brazil and Korea.
Vincent also is interested in the workshop on NGPON.2 also being held on Monday and being led by Frank Effenberger of Huawei Technology. Verizon is committed to having its next passive optical networking (PON) offering for FiOS ready by 2015. “We are starting to whittle down the number of technologies from a dozen to six or so,” he says. OFC provides Vincent the forum to gather the information he needs to help make those critical decisions.
For Paul Baniewicz, Alcatel-Lucent’s vice president of optical for the North American region, the conference offers the opportunity to hear from the vendor’s most significant customers. “When they have to put what they need into some sort of format and they are sitting side by side it becomes clearer,” says Paul. “They are all looking to make a number of technology jumps.”
Baniewicz does not think of OFC as only an optical components show. The systems vendor is making a major commitment to this year’s conference. “We will make announcements and will have equipment there,” says Paul. “This is the first time we are bringing in equipment.”
Joe Berthold, Ciena’s vice president of network architecture, wants to know “how broad the recognition is at the conference that coherent technology really simplifies dynamic reconfigurability of the photonic layer of the network."
In short, it is a place where the fiber industry gets updated. No one person even with today's Internet and search engines could be aware of everything that is occurring in optical transport. This is particularly true as technology continues to be pounded into the optical transport network at an astonishing rate.
Some other major themes at this year’s show revolve around 100 G, the next step after 100 G, coherent optical, flexible and intelligent networks, upgrading components such as reconfigurable optical add/drop muiltiplexers (ROADMs) and optical amplifiers, and photonic integrated circuits.
The ability to deliver 100 G in a manner that is competitive with 40 G and 10 G is a major focus of concentration. Cost-conscious customers have gone to the 10x10 MSA connection (10 lasers each operating at 10 Gbps) rather than the IEEE standardized 4x25 (four lasers each operating at 25 Gbps) because it is less expensive. A concern of Joe’s that he hopes will be addressed at OFCNFOEC 2012 is related to the high cost of the client interface for 100 G.
Paul Baniewicz agrees the industry at least for now is gravitating to 10x10 and says the per bit cost for 100 G is rapidly approaching what it is for 10 G. Even though a lot of 40 G gear is being sold now, Paul believes 100 G will come in strong.
“I look to shows like OFC to better understand what will happen after 100 G,” says Paul. Indeed, there is a lively debate about whether the industry should go to 400 G or 1 Tbps. 400 G conforms to the SONET/SDH 4:1 demarcation, while 1 Tbps is in line with the 10:1 Ethernet increases. Engineers are largely focused on the issue of spectral efficiency. He explains that 100 G can be delivered over a 50 G spectral efficiency and engineers have been able to deliver 400 G with spectral efficiency of 100 G, which is only a twofold increase, rather than fourfold. This potentially means a lower impact on the parameters of existing networks and so is advisable.
But that is far from being the last word. Proponents of hopping to 1 Tbps point to the struggles of going to 40 Gbps and note that it will be some time before 100 G reaches the level of use of what 10 G is now being deployed at. By that time, the technical issues should be fully addressed.
Coherent optical has been a topic that has been all the rage at recent OFCs. But while it is commonly accepted as being critical for transmitting 100 G over longer distances, some vendors have raised the question as to whether the extra cost makes sense for shorter distances, such as in the metro network or in shorter European hops. This is seen as being another area of lively debate, especially as coherent technology starts to fall in price. “Coherent transceivers will progress,” says Joe Berthold, “there will be simplified versions of coherent transceivers that will address markets that don’t need the highest performance.”
Another area of keen interest at the conference is building more intelligent and more flexible networks. Carriers would like to have as much of the network automated as possible. “If you look at the future type of transponders, they are going to be software defined,” says Paul. How well will a network be able to shift speeds to use smaller transmission rates for longer distances and then have the flexibility to send at larger transmission rates for shorter distances? “It’s all about getting the most capacity on the fiber,” says Paul.
Fiber optic system vendors have always been dependent on optical component suppliers to large extent in implementing new innovations in the network and that will be key focal point at the show.
While ROADMs have been a hot conference topic for several years, Joe says ROADM capability often can boil down to economics. “We actually introduced colorless, directionless and contentionless ROADMs some years ago, but the percentage of buyers that wanted all three was quite small,” says Joe. “We have had some customers that have studied it quite thoroughly. They may buy a ROADM that is colorless and coolerless but pass on the contentionless part because it is not worth the extra expense.”
As networks experience faster transmission rates going at longer distances they require more sophisticated optical amplifiers. The move is to go to Raman amplifiers although erbium doped fiber amplifiers still play an important role. A topic of conference discussion undoubtedly will be new hybrid amps such as those being pioneered by Israeli-based Red-C.
Photonic integrated circuits (PICs) continue to hold tantalizing potential for doing the job quicker, faster and cheaper. Joe notes that PICs for 100 G are starting to look “pretty attractive.” But he posits a challenge. “We would like to see them further integrated so they can be put into one inch square formats.”
Indeed, that in some respects is the challenge and excitement of OFCNFOEC 2012. We know one way of doing something. Can you do it a better way? Those that can will be tomorrow’s winners.
Posted: 21 February 2012 by
C. David Chaffee, Founder and CEO, Chaffee Fiber Optics
| with 0 comments