Centrality Based Geocasting for Opportunistic Networks

Opportunistic networks (OppNets) are a subclass of delay tolerant networks characterized by unstable topology, intermittent connectivity, and no guarantee of the existence of an end-to-end path to carry the messages from source to destination because of node’s mobility. For these reasons, the routing of data in such networks occurs in a store-carry-and-forward approach. Following this approach combined with geocasting as message transmission strategy, this paper proposes a geo-casting protocol for OppNets named Centrality based Geocasting for OPPortunistic networks (CGOPP for short). The proposed scheme combines uni-casting and flooding by first transferring a message to an individual geographic location, then intelligently flood the message to all nodes inside the area by using a degree centrality function. Simulation results are provided, demonstrating the efficiency of the proposed protocol against traditional geo-casting protocols for OppNets, in terms of average latency, delivery ratio, and number of messages forwarded, under varying number of nodes, buffer sizes, and TTL.