Based on bit rate alone, DWDM has a fourfold advantage even over the latest—albeit nascent—TDM option, STM–64/OC–192. To fairly compare the two technologies, however, we need to review and outline what would be an ideal technological solution for expanding network capacity. This has to be done in a broad sense, recognizing that there are instances in which TDM may offer a better solution than DWDM. Analyzing the alternative attributes and benefits of each approach would require a comparison of several key issues:
1. Compatibility with Fiber Plant. The majority of the legacy fiber plant cannot support high bit rate TDM. Earlier vintage fiber has some attributes that lead to significant dispersion and would, therefore, be incompatible with high bit rate TDM. Recently produced fiber— NZDSF, for example—is flexible enough for the latest TDM equipment, but it is expensive and may limit the ability of carriers to migrate to the greater bandwidth available through DWDM at STM–16/OC–48 rates.
2. Transparency and Interoperability. The chosen solution must provide interoperability between all vendors’ transmission equipment, both existing and new. It must be vendor independent and conform to international standards such as the proposed ITU channel spacing and be based on the Open Systems Interconnection (OSI) model. Furthermore, it must be capable of supporting mixed protocols and signal formats. Some commercially available DWDM systems provide such transparency and can be used with any SONET/SDH bit rates, as well as with asynchronous/PDH protocols.
3. Migration and Provisioning Strategy. The best solution must also offer the ability to expand. It must be capable of supporting differing bit rates and have channel upgrade capability. It has to be a long-term solution and not just a short-term fix. TDM systems already are reaching their technological barriers and STM–64/OC–192, although rich in capacity, may represent a practical limit that could only be superseded by DWDM.
4. Network Management. A properly engineered solution should also support a comprehensive network element management system. The solution must meet international standards, interface with the carrier’s existing operating system, and provide direct connection for all of the network elements for performance monitoring, fault identification and isolation, and remedial action. Sophisticated and reliable network management programs will become increasingly important to deal with the increased complexity and expanded capacity that will be unleashed through migration to optical networks.
5. Technical Constraints. The systems deployed must be able to resolve some of the outstanding technical issues present in current lightwave transmission systems. For example, signal dispersion compensation, filtering and channel cross talk, nonlinear four-wave mixing, and physical equipment density are some of the more common problems. Ideally, an optimized system level architecture that provides a coherent and unified approach should be chosen over one that involves the acquisition and deployment of components on a piecemeal and uncoordinated basis.
