An Analytic Approach to Efficiently Computing Call Blocking Probabilities for Multiclass WDM Networks

For all-optical WDM networks that provide multiple classes of service, we present a methodology for computing approximate blocking probabilities of dynamic routing and wavelength assignment policies. Each service class is characterized by its resource requirements (number of wavelengths needed for a call) and expected call holding time (or subscription period). Under the wavelength continuity constraint on lightpaths and loss network formulation, we develop fixed-point approximation algorithms that compute approximate blocking probabilities of all classes. We then apply them to the random wavelength assignment policy for the following wavelength routing policies: fixed routing (FR), least loaded routing (LLR) and fixed alternate routing (FAR). Simulation results on different network topologies and routing policies considered demonstrate that the simulation results match closely with the blocking probabilities computed by our methods for different multiclass call traffic loading scenarios.