Admission control strategies in transient network states

This paper deals with call admission control (CAC) schemes based on the knowledge of the topology, configuration and load of the network. While most research has focused on admission control in normal conditions, we consider the particular case of "irregular admission control" which takes place just after a change in the network has occurred, making the network knowledge of the entity providing admission control obsolete. An overall model is presented, consisting of different states, events for the transitions between those states, and actions within the states. A method to detect network changes is proposed. Then, different admission control strategies to be applied during the following transient state are discussed. Finally, some approaches are described for mapping the connections onto the new topology after the network stabilization.