The parallel use of multiple channels in a WDM star network means that too many packets may simultaneously arrive for the same destination station, necessitating the implementation of a destination-conflict-resolution function somewhere within the network. This thesis considers explicitly, the placement of the destination-conflict-resolution function which specifies the location(s) where it should be performed, and when it should be performed. Traditional placements in which the function is located at all user stations and performed either before packet transmission (using the request-schedule-then-transmit principle) or after a destination conflict has been detected (using the detect-and-retransmit-if-lost principle), is compared with a central placement in which only one central station located at the entrance to the star coupler is responsible for detecting conflicts and re-scheduling the arrival times of “otherwise lost” packets whilst they are en route to their destinations, so that they arrive when their destinations free to receive them. The networks are evaluated considering their delay and throughput characteristics, the computational complexity of their protocols, and their hardware demands. All numerical results were produced using AKAROA an object-oriented parallel simulation package developed by us for automated precision control of steady-state estimates and automated parallel execution of quantitative simulations. The results presented suggests that significant performance improvements are achievable with the central placement since destination conflicts are resolved without having to retransmit packets nor waiting until the end of a request-broadcast-and-schedule phase before a given packet can be transmitted. The central station works with Space Division Multiplexed signals, just before they enter the star coupler. Its implementation is therefore simpler than when all stations are charged with this task, each of which has to attend to multiple WDM channels. Only “otherwise lost” packets are buffered so the network has low buffer memory requirements.