(This review has been adapted from the OIDA Optical Communication Roadmap and the OIDA Newsletter. For more details, contact OIDA, 2010 Massachusetts Avenue, N.W., Suite 200, Washington, DC, 20036-1023; Tel: (202) 785-4426; Internet: http://www.OIDA.org).
Last year, the Optoelectronics Industry Development Association (OIDA) and the Defense Advanced Research Projects Agency (DARPA) organized a workshop to identify important opportunities in the rapidly growing optical communications industry. The workshop, which was held May 18-19, 1998 at the Washington Hilton Hotel, included forty-nine experts from industrial, academic and government agencies. It should be noted that while this workshop was tasked to address systems’ requirements for components, individual component technology is to be studied more extensively at a workshop this year, and will serve to round out the activities planned to update the first Optical Communications Roadmap published in 1994.
Two important resources, a ‘strawman’ technology study and a study on markets for products and systems, were provided to participants to serve as a basis for discussion, and were incorporated into the study. The background technology report, provided by Peter Kaiser (Kaiser Global Consulting, NJ), noted the convergence of services in the information infrastructure is dominated by the growth of the Internet (see Figure 1). However, it is precisely this convergence that provides a unique opportunity for the optoelectronics community to develop the necessary high capacity flexible and reliable systems required. The latter resource, provided by Richard Mack (KMI Corp.) also identified the rapid growth of the Internet as a motivating factor for upgrading networks that support a strong communications industry.
In the past, the predominate traffic on the network was voice traffic. Predictions on the relative network traffic (see Fig. 1) suggest that within a few years, Internet traffic will surpass voice traffic by 2001 and that voice will only occupy ~ 1% of the total traffic by 2005. It should be noted that the total capacity is obtained by assuming a factor of 8 between the total average traffic and total capacity. This evolution in the composition of network traffic suggests that there will be an accompanying evolution of the traditional hierarchical network topology of circuit switching to that of a flat network topology for packet switching that favors data and Internet traffic. Key technologies that were identified by the workshop participants were single tunable lasers to replace multiple fixed wavelength lasers, ultrawide bandwidth fibers and optical amplifiers, and large capacity optical cross connects and routers for handling multi-terabit information rates on a single fiber. Given the multiple wavelength architectures dominating high capacity networks, the issues of how to manage hundreds of wavelengths efficiently and effectively become paramount.
Owing to the rapidly growing data traffic market and IP based multimedia services, the evolution to a packet switched, flat networking topology is inevitable. DARPA’s Next Generation Internet program, and Gigabit Ethernet technologies at 10 Gb/s, will serve as a motivation for bringing fiber to the desktop. This however, will require very low cost transceivers, fiber connections and installation. The emergence of unshielded twisted pair technology (UTP-5) will provide for drastic cost reduction of 10/100 Mb/s network interface cards, and as a result may make it difficult for fiber to reach the desktop. In high performance business environments, however, network managers may wish to deploy fiber immediately to avoid costly upgrades and delays at some later date.
For access networks, the bandwidth bottleneck within the last mile must be eliminated in order for multimedia services to be provided to customers. Fiber to the home is most attractive optoelectronic solution, however, the cost of optical network units have to be reduced before widespread deployment can be expected. Many of the issues associated with cost reduction is being examined by the Full Service Access Network Consortium. The key approach to cost reduction is in standardizing specifications so that multiple vendors can focus on developing new cost effective approaches to access network technologies and devices.
Estimated growth of traffic on the core network. Total capacity is 8x the average traffic. After the OIDA Optucal Communications Roadmap.