The demand for bandwidth has made optical components which increase fiber throughput, one of hottest areas of research and business alike. The field requires highly specialized and interdisciplinary know-how, such as that found in Switzerland. The high number of new Swiss start-up's focusing on the field leads to hope that Switzerland will play a significant role in this attractive emerging market.
Andros Payne*
Like other suppliers of enabling technologies, few people have heard of companies like JDS Uniphase, SDL, Sycamore, New Focus or Corvis. However consider that JDS Uniphase has approximately twice the market capitalization of General Motors but with only 1% of the giant auto maker's sales. Corvis went public in June 2000 with a market capitalization of over US$30b…and no sales. Who are these companies? What makes them special? And why do they command such high valuations?
Simply put, each of these firms is a part of the emerging optical networking industry. Point-to-point fiber-optic transmission has been used for over twenty years to transport large volumes of data over long distance fiber "pipes". The first commercial fiber-optic applications were for long, undersea links where both traffic volumes and the cost of laying fiber was high. However today, as a result of fundamental changes brought-on by the internet, fiber-optic transmission is now going through a renaissance which will have an impact on daily life around the globe.
In contrast with normal electro-magnetic transmission over copper or metallic conductors at Megahertz and Gigahertz frequencies, light transmission takes place at the orders of magnitude higher Terahertz frequencies. At these extremely high visible light frequencies, bandwidth is essentially unlimited and allows the transmission of unusually high volumes of data. Siemens and Corning recently demonstrated the transmission of over 7 terabits per second over a single fiber, enough for over 100,000,000 simultaneous telephone calls! Besides the glass fibers themselves, the electronic boxes which create, amplify, regenerate, switch and receive tiny light pulses are critical drivers of transmission system performance and bandwidth capacity. This interdisciplinary field between optics, electronics and telecommunication is know as opto-electronics.
The internet is also fundamentally changing the structure of telecommunications networks architectures. Prior to the internet, most traffic consisted of voice communications which were generally local connections. Unlike voice communications however, internet information is made up of digital data stored in servers at all corners of the globe, far from an web surfer trying to access it. This means that the internet and data communications will shift the proportion of traffic from the local networks to long-distance fiber-optic backbones. The rapid growth in the number of internet users has already driven an expansion of backbone network capacity. This will likely accelerate as advanced internet applications, such as video downloading, become more available. The bandwidth per surfer, not just the number of surfers, is rising quickly and placing increasing demands on fiber-optic capacity.
Despite the internet's growth, one critical piece of the bandwidth demand puzzle has, until recently been missing. No matter how exciting the internet content was, only the privileged few with a broadband local access connection could fully enjoy it. All the rest had to satisfy themselves with frustratingly slow web pages that could not be used to view video. That will now change. A host of new broadband local access technologies are being rolled-out across Europe, North America and Asia. Swisscom, Cablecom and hundreds of other network operators around Europe have quietly been installing broadband access to most households. The most popular technology among incumbent operators is ADSL (Asynchronous Digital Subscriber Line) which utilizes their existing copper wire infrastructure. However this is just one of many technologies which are being competitively implemented in parallel. High-speed cable modems and fixed wireless broadband are also being commercially rolled out. In addition, next generation mobile networks will demand more bandwidth than their slow-moving GSM cousins. Combine these broadband access options with a logarithmically growing number of internet-ready consumers, and you have a recipe for massive bandwidth growth in the future. Leading market research firms estimate that backbone traffic will grow at around 80% per annum through 2005.
Opto-electronic companies are aggressively developing components to alleviate constrictions not just in long-distance transmission but also in switching. Currently, packet switching is done electronically in a Cisco router. This requires incoming long haul optical signals to be converted to electrical signals before being packet switched, converted back to light frequency and then rerouted out over another fiber-optic line to the destination. This cumbersome approach will have to go. An all optical router that is perhaps five to ten years in the future will someday replaced it. It is no wonder that Cisco is currently investing heavily in opto-electronic component manufacturers.
The question remains - why are opto-electronics companies so valuable? The answer is only partly explained by booming demand for bandwidth. Unlike hundreds of thousands of internet start-ups sprouted by students on a shoestring, competing in the opto-electronics field demands a very unique set of highly specialized skills not easily found outside the quantum physics, applied optics and electrical engineering circles of elite universities and corporate research parks. The recent boom in demand has caught this once sleepy and academic industry off-guard. A fragmented set of players at the component level, few technical standards and lack of mass production skills are the typical signs of a market in infancy where first movers still have an advantage.
Could Switzerland play a key role in this emerging market of the future? Strangely, the answer is a resounding YES. A host of opto-electronic start-ups as well as established players already call Switzerland home. Among the promising new Swiss firms are GigaTera which focuses on high-speed, long-haul transmission, Sercalo which is working on all-optical switches and Luciol which makes fiber testing equipment. All are outgrowths of local university research. JDS Uniphase is the world's largest opto-electronics competitor. Although a Silicon Valley based firm, it has over 300 highly trained employees at its Zurich facilities. Switzerland has more opto-electronics activity per capita than perhaps any other European country except for Scotland. Not a bad starting point.
Switzerland's opportunity however lies in having the unique skills to solve the opto-electronics industries' greatest current problem - learning how to mass production minute, high-precision components. Highly trained operators still assemble most opto-electronic components by hand. Micron-level tolerances, individual adjustments and rapid design cycles have kept manual assembly the industry norm. There is a need for precision machines and production processes to automate opto-electronic component manufacture. A few Swiss machine manufacturers like Sysmelec in Neuchatel have already begun to focus on this emerging new market.
However, despite a promising start for a few entrepreneurs in the area and an outstanding academic and industrial infrastructure, the Swiss response to opto-electronic opportunity has been ad-hoc and arms-length. The key question is whether government and industry leaders can coordinate Switzerland's uniquely relevant electronic, optics and micro-mechanics expertise to forge a leading position in one of the most promising industries of the 21st century.
* The Author is CEO of GigaTera Inc., Zurich (Switzerland).
December 12, 2000
You find the original of this articel on NZZ Online under:
http://www.nzz.ch/2000/12/12/qo/page-article6XE9F.html