For example, carriers will still have to design for all possible
208 Chapter 8
link budgets (e.g., EDFA per 80??“120 km distances) and handle dispersion effects at
higher 10 Gbps bit rates (e.g., DCF coils on selective spans).
Mesh Switching Mesh fiber networks represent the next topological progression from
OADM rings. For example, most long-haul backbones are of a mesh nature and large
metro domains also comprise a meshed interconnection of rings. In order to provision
wavelengths over these topologies, generalized optical cross-connect switch (OXC)
nodes are needed to handle increased fiber connectivity (refer back to Figure 8.3, earlier
in the chapter). Today, nearly all such designs use SONET/SDH digital cross-connect
(DCS) [5] fabrics with full regeneration of all wavelengths, which is very costly.
In order to increase scalability and transparency, all-optical fabrics have also been
considered, such as two-dimensional MEMS. Nevertheless, these systems face a host of
cost and complexity limitations and have seen very minimal traction with carriers. For
example, current MEMS fabrics are limited to smaller 16 ?— 16 sizes and require costly
multistaging to scale port counts, further increasing loss and crosstalk (Clos, Banyan,
etc.
Pages:
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521