pavnet paves the way for plastic optical fiber

A new darpa-sponsored effort should help overcome the barriers to the use of plastic optical fiber in telecommunications applications.

Ever-growing bandwidth requirements and the desire to obtain this bandwidth at low cost are driving an increased interest in plastic optical fiber (pof). These increasing data requirements also drive susceptibility to electromagnetic and radio-frequency interference in copper-based systems. pof technology is immune to such interference, as is glass fiber. Yet pof has a demonstrated ease of termination and serviceability that glass does not possess. Because of its large core size, it also has a much lower component cost--i.gif., for connectors and transceivers--than glass. This means potentially lower installed and life-cycle costs. Because of the possible benefits of pof, industry and government have become interested in the development of this technology. Consequently, a consortium was formed to advance the development and use of high-speed plastic technology.

Much of the research into refining pof technology for communications applications has taken the form of partnerships between government and industry. For example, in 1993, the Advanced Research Projects Agency (now the Defense Advanced Research Projects Agency, or darpa) awarded a contract to Packard-Hughes Interconnect, Boston Optical Fiber Inc., Boeing, and Honeywell Inc. to develop high-speed, low-cost pof technology under the High Speed Plastic Network (hspn) program. After two-and-a-half years, the team developed the first commercially available graded-index plastic fiber with a bandwidth greater than 3 Gbits/sec when measured at 100 m. The hspn team also demonstrated the use of this technology in aviation local area networks carrying both Fiber Distributed Data Interface and Asynchronous Transfer Mode traffic, as well as in a Fast Ethernet network and an automotive satellite link.

As a result of achieving their goals, the team, in conjunction with Lucent Technologies, was awarded a contract in 1997 by darpa to further exploit pof technologies for use in harsh military and rugged commercial environments. This program is called Plastic and vcsel Network (pavnet). The cornerstone of the pavnet program is the development of new pof material systems and manufacturing processes.

In the course of its work, the pavnet team will develop a graded-index plastic fiber that is wavelength-independent from 500 to 2000 nm with very low attenuation-- <60 dB/km. In addition, the fiber`s operating temperature range will increase from today`s 85C to 125C. These developments will facilitate the use of pof for longer distances and will allow the use of longer-wavelength light-emitting diodes and lasers. The use of wavelength-division multiplexing promises to further increase channel capacity as well.

Besides researching the new fiber, the pavnet team will also focus on the development of a new connection system, termination techniques, and electro-optic modules that incorporate low-cost, reliable infrared vertical cavity surface-emitting lasers (vcsels). These components will be used to demonstrate the new technology`s operability in military and commercial applications.

Technologies developed under pavnet will provide significant benefits to industry and the military beyond those demonstrated in the program. Among the applications are office networks, high-speed Internet access, industrial control, residential broadband networks, commercial aircraft, intelligent transportation systems, and military applications such as tanks and helicopters.

Meeting challenges for telecom

Four of the major stumbling blocks for pof technology in meeting the needs of telecommunications networks have been a lack of bandwidth, high attenuation, temperature performance, and cost. The hspn program answered the bandwidth and cost issues. Two of pavnet`s major challenges are temperature and attenuation. The root of these problems in previous generations of pof has been thought to be the polymethylmethacrylate material used in fiber construction. The pavnet effort will investigate the use of a different material--a fluoropolymer--which promises to exceed the program`s temperature goal of 125C. Meanwhile, the dopant materials integrated with the base material should provide graded-index fiber losses potentially less--and possibly much less--than 60 dB/km.

The challenge in the area of transceiver design will be to leverage existing infrared vcsel technology for use with large-core pof. Present transceivers are optimized for small-core glass-based fibers. By leveraging existing vcsel technology, the team can take advantage of such lasers` low cost and maturity.

The pavnet team is also working to reduce the cost of pof interconnection by determining the commonality between automotive and aircraft applications. Applying the lessons learned from this investigation will help realize economies of scale, which in turn should encourage high-volume use and thus lead to reduced costs. At the same time, each interconnect design will incorporate the packaging necessary to meet the specific requirements of its intended environment. These requirements include environmental conditions, architecture, and system performance.

Technology demonstrations

The interconnects developed under the program will be integrated into the demonstration and test platforms that Lucent and Boeing will provide as part of the pavnet contract. Lucent will oversee a demonstration centering on a telephone central office switch environment, while Boeing will conduct a demonstration of pavnet technology in a dual-use military/commercial aircraft application.

The pavnet work is particularly promising for intrasystem applications such as interconnection within large telecommunications systems and computers. Tackling this issue is the primary reason that the transmission business unit of Lucent Technologies is participating in the pavnet effort. The cost of interconnection within large digital terminals--such as digital access and crossconnect systems--is a significant amount of system cost. It is expected that pof will reduce the required precision and hence, the cost, of interconnection hardware.

The central office demonstration should show how the different elements of the pavnet effort can work together to make pof an even more viable option for telecommunications applications. For example, the bandwidthdistance product of today`s step-index pof is inadequate to meet the requirements of such an application. Fortunately, the graded-index pof mentioned previously promises to fill this role. Today`s transmission terminals use a combination of metallic interconnection and multimode glass-optical-fiber interconnection. The use of this advanced graded-index pof is expected to produce significant cost reductions for the interconnection hardware within large transmission terminals.

Meanwhile, the wavelength-independent graded-index fiber will allow the use of high-reliability, high-speed trans-ceivers based on infrared vcsels operating in the range of 850 to 1300 nm. The distance requirements in such applications are generally under 30 m (although distances to 100 m may be desirable) and the bandwidth requirements today range up to 622 Mbits/sec. In the next two to three years, this will probably increase to 2.5 Gbits/sec over the same distances. The expected pavnet technology base of advanced graded-index pof, vcsel-based transceivers, and low-cost connectors is expected to encourage the design of powerful new switching and interconnection technologies based on the extensive use of serial optical links.

For its part, Boeing`s portion of pavnet will emphasize cost-effective, high-bandwidth pof interconnects for distribution within the cabin and airframe of commercial and military aircraft. Despite the benefits of glass optical fiber in such applications, the cost of flight-qualified glass fiber data links has limited the technology`s widespread use in cabins to network backbones. Simple field terminations of plastic fibers may reduce life-cycle costs enough to justify significantly increasing the fiber content and bandwidth of mixed multimode fiber/copper-cable harnesses. Simple, high-density, easily reconfigurable ribbon cable harnesses with breakouts distributed from centralized switches will permit orders of magnitude bandwidth increases relative to today`s shared bus daisy chain interconnects.

The pavnet program will conclude in 2000. At that time, the pavnet team foresees that this newly developed pof technology will be the "2000+" technology of choice for commercial and military short to medium-length applications. u

Michael Orr is product manager of fiber optics at Packard-Hughes Interconnect, Irvine, CA.

Get Answers--Fast!

To get the quickest response to your question, contact the appropriate person listed below.

Reprints and Bulletin Board

Kathleen McIntosh

Tel: 603-891-9203

E-mail: [email protected]

Subscriptions

Ruth King

Tel: (918) 832-9349

E-mail: [email protected]

New products, calendar