For anyone who has been following the stock market of late, or even high-technology business news, it should come as no surprise that component makers in the field of optical fiber communications are scrambling to meet the increasingly challenging demands of network designers. Fittingly, this scenario is also reflected in the technical program of the 25th annual Optical Fiber Communication (OFC) conference, scheduled for March 5-10 at the Baltimore Convention Center (Baltimore, MD).
"The concepts are there but not all of the components are available with the proper specs," said Wayne Sorin, principal project scientist at Agilent Laboratories (Palo Alto, CA) and OFC technical program chair for presentations concerning components. "It's just a challenge. So I think in the component area, they're working hard to try to meet the requirements that these future networks are going to have."
One of the most active areas in terms of meeting this challenge at OFC 2000 is that of L-band amplifiers (see figure). More than 40% of the papers submitted to the subcommittee on Fibers, Fiber Amplifiers, and Propagation were related to extending the wavelength range of amplifiers, Sorin said. Of course, broader bandwidth is just one aspect of the evolving balance in fiber communication growth between wavelength-division multiplexing and time-division multiplexing, which is also reflected in the scope of technical presentations this year.
"Generally there is still a debate about the best way to make use of the bandwidth in a fiber," Sorin. "You can go to higher bit rates, like from 10 to 40 Gbit/s, and there are a lot of people working in that area. The other way would be to slice up the spectrum into smaller windows, like windows of 100 GHz." Ultimately cost issues are likely to play a major role in deciding the balance between the two.
"Is it cheaper to go to the higher bit rates or can the component people make these filters cheap and reliable and stable enough that you can make these slices narrower? I guess whoever can do it more economically will determine which way it actually ends up," Sorin observed.
Extending erbium-doped fiber amplification into the L band will be a major focus of technical presentations at OFC 2000. The above schematic, to be discussed by Lucent Technologies (Holmdel, NJ) researchers during a Wednesday afternoon presentation at 2 p.m., describes an experimental setup for measuring dynamic behavior of L-band EDFAs.
In addition to high interest in the L band, a few presenters have begun to explore the S band at the short end of the spectrum, according to Irl Duling, chair of the subcommittee on Fibers, Fiber Amplifiers, and Propagation, as well as a lead research physicist at the Naval Research Laboratory (Washington, DC). A number of papers in the same category will also explore hybrid amplifiers that split the data coming in at the input into L and C bands or that incorporate Raman amplification.
Presentations concerning dispersion compensation are taking on more complexity also because it's no longer enough to just shift the dispersion to where you want it in order to make a dispersion-compensating fiber, Duling said. "Now they want it to have the opposite sign from third-order dispersion so that you can also compensate for the slope of the dispersion curve over broader ranges."
The new, new thing
Of course, photonic bandgap crystals (holey fibers, in this case) will be covered at OFC. Much of their allure may lie in the potential of sparking a new paradigm in the field, Sorin said. "This is one of these possible new things that might blossom into something in the future," he said. "In the past the EDFA [erbium-doped fiber amplifier] was really a big kick for OFC, and then fiber Bragg gratings were the hot thing for a while. There haven't been any new twists lately," he added. "And so holey fibers may be a new thing that will cause people to think differently and open new possibilities."
Jay Wiesenfeld, a technology consultant at AT&T Labs (Red Bank, NJ) and chair of the subcommittee on optoelectronic devices is looking forward to a potpourri of new and interesting work in lasers, receivers and photodiodes, nonlinear applications of laser and microelectro-optical amplifiers, optical switches, MEMS devices, and micromachine devices.
"We continue to see progress in the basics of this field, which are the lasers and receivers," he said. "There's a lot of work in optical signal processing, which is really enabled by the increasing capabilities of monolithic integration allowing fairly complicated photonic integrated circuits. MEMS micromachining work continues on many fronts, producing a variety of devices. And optical switching technologies are an important driver."
Tuning in
In the laser area one of the main issues is WDM sources, either wavelength selective or tunable or some combination, he said. Strong papers will be presented on distributed-feedback (DFB) laser arrays. "There are a number of lasers packaged monolithically on a chip at different wavelengths, and tuning is coarsely done by selecting which DFB to choose and then it is finally done by temperature tuning," he said. "This is an idea that has been worked on for a few years, but what's really good is these devices are now achieving significant output powers in the 7- to 10-mW range." Wiesenfeld also cited progress in multifrequency laser devices with integrated array waveguide grating routers on-chip combined with various amplifier sections to yield a wavelength-selective source.
In terms of receivers, there's a push now toward higher speed, with a lot of 40-Gbit/s components coming on line and maturing. In the similarly burgeoning area of optical switching, talks will range from alternatives to optical switches (an invited talk on electronic fabrics for optical crossconnects) to both invited and contributed talks on a type of optical switch based on internal reflections in a planar lightwave circuit and mediated by the presence or absence of index-matching fluid.
In the area of micromachines and MEMS, one invited talk covers novel micromechanical network elements, a contributed paper describes a novel micromechanical optical attenuator, and other talks describe micromachine space switches based on optical micromachines.
Hassaun Jones-Bey