Figure 1. At the heart of Lucent Technologies' WaveStartm Lambda Router, an all-optical lightwave-routing device, is an array of microscopic mirrors, each able to tilt in various directions to steer light. The micro-mirrors route information in the form of photons to and from any of 256 input/output optical fibers.
The vast expansion in the use of the Internet over the past two years resembles the upsurge in tourists traveling to an increasingly popular resort. Traveling along the main highway to the resort at the start of a summer weekend, the stream of tourists experiences a series of bottlenecks whenever the six-lane highway narrows down to four lanes to cross bridges or pass other obstructions. Internet users face similar kinds of bottlenecks. These take the form of hub regions that must switch pulses of light moving rapidly and efficiently along optical fibers onto new routes. At those hubs, the system must convert the light pulses into electronic packages to permit them to travel through the switch. Once rerouted, the packets are converted back into light.
The conversion and reconversion slows down the entire process. David Bishop, director of micromechanial research at Lucent Technologies' Bell Labs, said that any sort of electronic system at the end of an optical fiber will inevitably cause a traffic jam because it lacks the capacity to deal with the fiber's bandwidth. So far, electronic switches have managed to keep up with increasing bandwidths; but as the amount of data sent over optical fibers is increasing twice as fast as the capacity of electronic switching, that situation won't last much longer.
Figure 2. One of an array of 256 microscopic mirrors, each the size of the head of a pin, tilts to steer lightwave signals from one optical fiber to another in Lucent Technologies' WaveStartm LambdaRouter, invented at Bell Labs.
The need for fast, reliable optical switches will only increase as technologies such as dense wavelength division multiplexing (DWDM) come into play. DWDM enables a large number of different data streams to travel along a single optical fiber. Routing those streams in the correct directions at hubs demands a technology more sophisticated than an optical-electronic-optical system can provide.
The obvious solution is to avoid the transitions between optical and electronic transmitters by using optical switches. Lucent and several other companies are working hard to develop optical switching devices, largely through the application of microelectromechanical and micro-optoelectromechanical systems (MEMS and MOEMS). "As we enter the new millennium, a strong demand will be created for optical switching for optical networking, telecommunications, and wireless technologies," said Edward Motamedi, executive director of Revoltech Microsystems and chair of several SPIE conferences on MEMS and MOEMS. The solution is far from monolithic. Competitors in the race for an effective optical switch have developed technologies that range from microscopic mirrors to miniature bubbles to liquid crystals. Some switches, notably those built by Lucent (see figures) and Optical Micromachines Inc., have already made their commercial debuts.
Mirror-based switches have drawn the most interest so far. Within the past 18 months, at least three companies have developed optical switches based on micromirror technology. Early last year, for example, Lucent's Bell Labs built what it called "The world's first practical optical switching technology using MEMS." The device is based on a tiny pivoting bar with a gold-plated mirror at one end that fits in a tiny space between two hair-thin optical fibers lined up end to end. When the switch is in the off position, the mirror rests below the cores of the two fibers, permitting light signals to travel across the gap from one core to the other. However, voltage applied to the far end of the bar helps to lift the mirror between the fibers, where it reflects incoming light rather than permits its passage. "Our optical switch shows the potential for MEMS to be a disruptive technology -- one that changes the paradigm for an entire industry," said Bishop.
Another example of Lucent's micromirror technology, its WaveStar LambdaRouter, uses 256 miniature mirrors, each no larger than the head of a pin, to steer light signals from one optical fiber to another. This promises the ability to direct traffic through optical networks 16 times faster than is now possible.
Early this year, Xros, Inc. announced "the world's highest capacity optical cross-connect system for open optical networks," based on its own micromirror technology. The system uses two facing 6 X 6-in. arrays, each containing 1152 mirrors made out of pure silicon. Computer-controlled electrical signals can tilt each mirror in several dimensions, allowing the system as a whole to direct any incoming light beam (or light path) in more than 1000 directions. That capability provides optical networks with extreme redundancy, in case of breakages in the optical fibers.
Optical Micromachines Inc., meanwhile, has used its new line of optical switches since February to route live traffic over an optical network in Oakland, California. The company said, "This represents the first time microelectromechanical subsystem-based switching systems have been deployed to carry live data traffic." Like Xros's switches, Optical Micromachines' are micromachined mirrors fabricated on silicon chips.
Other technologies rely less on MEMS and MOEMS. Agilent Technologies, Inc., a company spawned by Hewlett Packard, uses bubbles to shuffle light packets between tiny waveguides made of glass. Corning Inc. and Chorum Technologies Inc. rely on liquid crystals to shift light beams from one path to another. Nanovation Technologies, meanwhile, is developing a hybrid technology that integrates MEMS-based optical switches with silica-on-silicon waveguides on a single chip.
One fact is common to the first generation of optical switch technologies: Whatever current problems they solve, they will soon be outdated. For while the switches effectively transfer packets of light from one path to another, they can't figure out the directions that each incoming light packet should take. That requires the ability to read rudimentary messages that the light packets carry. At present, only microelectronic systems have that ability.
Few MEMS researchers doubt that optical systems will eventually catch up. By mimicking optical circuits, the researchers expect to devise intelligent optical networks that integrate MEMS and waveguide technologies. The ultimate goal of Internet transmission mediated entirely by light is coming into sight.
A former science editor of Newsweek, Peter Gwynne is a free-lance science writer based in Sandwich, MA.