Equipment manufacturers are actively evaluating tunable laser technologies and are ready to take advantage of these products when they become available, write Lisa Huff and Christine Mulrooney of Communications Industry Researchers (CIR).

To achieve the vision of a programmable network, equipment providers are actively considering the use of tunable components, and, more specifically, tunable lasers. These devices allow a level of flexibility that is unavailable today, and have obvious benefits for optical networking systems. In DWDM networks, for instance, tunable lasers can theoretically reduce the number of lasers required to support 128 different chan-nels from 128 lasers with discrete wavelengths, to 8 lasers with 16 tunable wavelengths. Instead of having to maintain a full inventory of spares for each channel or port, service providers would be able to stock a few tunable transmitters that could be used to replace failed parts. Sparing is just one potential application of tunable lasers. More exciting is the prospect of remote service provisioning of the network. This would allow service providers to remotely change the wavelengths utilized by a specific customer to provide "bandwidth on demand". The potential cost and performance benefits of such an approach are enormous. At present, however, the reality is that tunable lasers are still a long way from realizing their full potential. While considerable growth is expected in this marketplace, most of the growth will not materialize for a few years, by which time tunability will have rapidly become a standard feature of many telecommunications lasers. Today s tunable reality At this time, there are only two uses for tunable lasers. The main use is in deployed systems, where tunables are used as temporary replacements for lasers that have failed, in order to keep the equipment running. The tunable laser is replaced with a fixed wavelength laser when the correct one is obtained. Tunable lasers are also being used in performance monitoring equipment to monitor several wavelengths at once. Our research indicates that most equipment manufacturers have been slow to deploy tunable laser technologies. There is general concern about the robustness and long-term reliability of tunable lasers. Today s products have shown good performance but only in the lab environ-ment, leading equipment manufacturers to believe that tunable lasers are not ready for broad implementation. Nonetheless, most are actively evaluating tunable laser technologies and are ready to take advantage of these products when they become available. Building the perfect laser For the tunable laser market to reach its full potential there must be some resolution of the numerous technological issues that currently exist. In particular, the market will have to determine the importance of the various trade-offs that are implicit in the tunable laser technologies now being considered. Equipment manufacturers and component suppliers alike are trying to obtain the perfect tunable laser using a variety of technologies, each with its own advantages and disadvantages (see ). In particular, there is a trade-off between wide tunability and optical output power for most device types. Many tunable laser manufacturers are arduously working at improving the reliability of their devices, and all are start-ing to incorporate integrated wavelength lockers in order to improve stability of the device at each wavelength. Improvements in manufacturing process controls are being realized, and most component manufacturers are expecting to be in volume production of qualified products by the end of 2001. Among these are Nortel High Performance Optical Components, New Focus, Agility, Alcatel Optronics, ADC, JDS Uniphase, Marconi Optical Components, Iolon and Corning. Fujitsu claims it is already in volume production with its devices, and that they are being incorporated into Fujitsu DWDM equipment. Nortel has addressed the stability issue by designing its tunable transmitters with integrated wavelength lockers and closed feedback for continuous tuning over the life of the device. Many other manufacturers are starting to use similar techniques in order to assure equipment manufacturers that tunable lasers are viable for system deployment in the long term. Future applications Although the technological issues outlined above will take some time to address completely, this process will eventually open up new applications for tunable lasers. Initially, these devices will be deployed as spares for fixed lasers that fail. But in order to take full advantage of the functionality of these tunable components, the topology and management of the network, including software, must change. Therefore, we see a relatively slow adoption of the devices over the next few years. Although equipment manufacturers are not currently deploying tunable lasers to their full functionality, they are excited about the prospects that these compo-nents hold. Once the reliability issues are resolved (and there is no doubt that this will happen), equipment manufacturers will use them for inventory reduction, performance monitoring and dynamic provisioning. Long-haul and metro markets The long-haul market will be the first to adopt tunable lasers. Historically, this market has been the proving ground for most new optical technologies, and is better able to bear the cost of these new technologies. In the short term, long-haul applications will utilize tunable DFB lasers, but will adopt DBR lasers and ECLs in the longer term. These latter device types offer broader tuning and higher optical output power. Because of its large size, the metro market is one area in which we see particular potential. However, before tunables can be widely used in the metro environment, costs will need to approach those of fixed wavelength lasers. Nonetheless, the metro market has a huge potential for growth and is already the focus of many tunable laser manufacturers. Since metro distances are relatively short, output power does not need to be as high as for the long-haul market. Cost is a huge consideration in this arena, so low cost technology, coupled with automated manufacturing, will be key for supplier success. The technology winners in the metro space are likely to be DFB lasers and VCSELs. Since DFBs are a proven technology, equipment manufacturers will find them easier to qualify and less costly than any of the other types. The inherently low cost of VCSELs will win them a share of the metro equipment business. The dynamics of the long-haul and metro markets will also influence the strategy adopted by companies developing tunable lasers. Concentrating on the metro market in the near-term is not a sound strategy unless the company has a revenue stream to sustain it for at least two years until the market develops. Equally, it could be a mistake to concentrate on widely tunable lasers if these do not have sufficient power to address the initial market for long-haul applications. Device types and markets Tunable DFB lasers are already starting to be deployed for sparing and replacements for fixed wavelength applications, and some equipment manufacturers, such as Fujitsu, are starting to incorporate them into their OADMs as light sources. Other tunable laser technologies have not reached the price points to be readily incorporated into systems. However, CIR sees a migration in the market over the next five years, from narrowly tuned DFB lasers to more widely tuned DBR lasers, ECLs and VCSELs, as the stability and price issues are resolved. The tunable transmitter market will remain fairly small until at least 2003 (see ). In 2001, the market forecast of $10.8 million consists entirely of narrowly tunable DFB devices. As the market improves, and applications such as wavelength routing and optical add/drop light sources start to develop, a $142 million market will emerge in 2003. Widely tunable lasers will become the majority of the tunable market in 2004. Sub-systems CIR believes that the future of tunable lasers will also include a transition of vendor focus from component level products to sub-systems. Eventually, most tunable laser manufacturers will sell transmitters, i.e. products at the sub-system or module level. This makes sense since the laser manufacturers are most familiar with controlling the tunability of their lasers, and have in most cases already developed transmitter sub-systems. The only problem with this is that if component suppliers have different tunable transmitters, equipment manufacturers would have to accommodate at least two different designs in order to have a second source. ADC and Agility recently proposed the obvious solution for the second sourcing issue, namely a multi-source agreement (MSA). This would allow component suppliers to develop tunable lasers using any technology but within a certain footprint and with standard control parameters. Such an MSA is likely to do more to enable the market than anything else. The success of this MSA is not only important for tunable transmitters and lasers, but also for the future of standardization of optical components. If this agreement is successful, CIR predicts that most other new optical sub-systems will be designed under similar agreements.