Vicinity routing in large scale networks

Efficient routing has been one of the most challenging topics in the communication networks field. Collecting the topology and network state information to every node has become a popular approach (termed a link state protocol). Applying this technique introduces a problem of a large amount of data and information updates. In large networks it leads to hierarchical division of the network into smaller clusters. Our solution takes a different direction. It defines a vicinity around each node that is updated with the local node and link information. Outside the vicinity an hierarchical yet flexible structure of the border node is defined. The route is calculated up to the nearest border and from this point a new calculation is made. This new architecture eliminates the need of dividing the network into clusters, in particular solving the inefficiency when such a partition is done manually