Placement of Wavelength Converters

The number of wavelengths available in a network is always limited due to the restriction of hardware structure of optical routers/switches. An important goal of the design of WDM networks is to use less wavelengths to serve more communication needs. There are two basic approaches to achieving the goal. The first one is to find the proper routing and wavelength assignment methods, that is the routing and wavelength assignment problem (RWAP) discussed in Chapter 3. In this chapter we will focus on the second approach, that is to use of wavelength converters. Existing study has shown that the more converters installed in a network, the less number of wavelengths is needed, given the same network load (the maximal number of channels over a link). In fact, by using enough number of wavelength converters at the network nodes, the number of wavelengths required can be made equal to the network load, this is most ideal situation that we can expect since the number of wavelengths required in a system is no less than the network load. A simple example of achieving this feature, called the Load-Wavelength Assignability (LWA), is to equip every node in the network with a wavelength converter. However, it is too expensive to do so, because this will not only increase the cost of network hardware but also the complexity of routing and wavelength assignment.