A probabilistic position-based routing scheme for delay-tolerant networks

Abstract

Observably, participants in realistic scenarios repeatedly navigate specific locations based on routine behavior, leading to inherently structured movement patterns. In this paper we propose a delay-tolerant routing scheme, called Probabilistic Routing with Minimum Proximity (PRMP), which aims to utilize prior movement patterns of peers to predict future probability of forwarding messages to a location proximal to a destinations home address. A source considers next-hop forwarding based on a probabilistic benefit-metric; which takes into account a nodes frequented trajectories and current position, its spatial distance from a destinations stationary home location and the probability of any of its immediate trajectories minimizing the spatial distance to the destinations home. Delivering a message to a nodes' home address in the network is synonyms to delivering mail to an individual's designated mailbox. The protocol avoids flooding completely in efforts to optimize use of network resources. Simulations of PRMP reflect low buffer occupancy at both high and low loads in the network. It also maintains resource optimization in varying node densities compared to two prominent DTN flooding protocols - Epidemic and PRoPHET.