Probabilistic routing using contact sequencing in delay tolerant networks

Abstract

Delay tolerant networks (DTNs) are characterized by intermittent node connectivity and unavailability of end-to-end paths for most of the time. Routing in such networks is a challenge because of these characteristics. Some existing routing protocols try to improve the delivery ratio by replicating some messages and hence perform poorly if the nodes have limited buffer space, or alternatively, if the number of messages is high. Some existing protocols also use the underlying movement pattern to improve routing efficiency. In many real life scenarios, movement patterns show repetitive contacts between nodes occurring in some time sequence. In this paper, we propose a probabilistic routing protocol that uses both the repetitive contacts and their time sequencing to improve routing. The protocol does not replicate any message and hence can work with low buffer size in mobile nodes. The novel feature of this protocol is the use of sequences of contacts which has not been used in earlier protocols. Simulation results show that the proposed protocol achieves high message delivery ratio with no message replication compared to some of the existing protocols.