MEMS Hang Tough in Optical Networking

What a difference a year makes. From the exhilaration of 2000 to the gloom of 2001, it's been a wild roller-coaster ride for the telecommunications industry. Yet, for MEMS companies pursuing this market, the march toward product commercialization continues relatively unabated. That's not to say that a few companies haven't experienced some growing pains.

Throughout 2000, at least four dozen companies announced plans to develop MEMS-based switches for use in the optical network of the future. A number of very high-profile acquisitions took place, and venture capital-based equity poured into these companies at levels never before seen in the MEMS industry. By the Optical Fiber Conference (OFC) in March 2001, the hype had reached a fever pitch.

At the same time, however, the market was beginning to crumble. Fortunately, for most MEMS firms, they were still in the development stage, and thus perhaps in a slightly better position to withstand the steep decline the greater telecom market was about to endure. Nevertheless, as with any burgeoning market, the news has been mixed. Some companies have moved forward into production despite market conditions, while others have experienced internal catastrophes.

The news that has caused the greatest stir to date is that both OMM and Onix Microsystems have recently retreated from their pursuit of 3-D MEMS switches, and subsequently laid off a considerable portion of their staffs. This has raised a lot of unwarranted questions about the ultimate workability of 3-D MEMS technology. To say that a technology is not feasible because two companies decided to refocus their businesses on products that were generating revenues is somewhat misplaced.

The fact of the matter is that this particular market segment is still some years away in terms of product maturity and, more importantly, real market need. In addition, this will be an especially competitive segment, and it will not realistically begin to take shape until late 2002/early 2003. In the meantime, there is positive movement forward. In May, Lucent Technologies announced the first sale of its WaveStar LambdaRouter (a 256×256 3-D switch) to Global Crossing.

As for 2-D MEMS switches, they're definitely gaining momentum. Certainly, revenues in this sector are not as high as they could be in a better economic climate, but shipping of commercial product is indeed occurring, and continues to grow. The litmus test will take place in the fourth quarter of this year, which is when the majority of companies anticipate ramping up production. While delays may be attributed to technology issues, the state of the telecom market and subsequent demand (or lack thereof) will have a far greater impact.

Perhaps the most extraordinary development over the past year is the diversity that has taken place in terms of product offerings. MEMS-based solutions have expanded from photonic (or purely optical) switches, to include variable optical attenuators, tunable filters and tunable lasers. As a result, there has been considerable market positioning among MEMS companies. For the most part, those who are developing switches (or lasers or filters) are focused solely on that type of product, although a few are looking to offer a diverse portfolio of MEMS devices.

Given the number of companies pursuing the development of MEMS-based products for use in optical networking, there will be plenty of news in the months to come with regard to who's doing what and why. More companies will change their business plans, others will be acquired, and still others will simply disappear. This is a market in transition and the name of the game at this point is finding a niche in which one can be successfully competitive.