HBPR: History Based Prediction for Routing in Infrastructure-less Opportunistic Networks

TLDR

Opportunistic Networks lack guaranteed end‑to‑end connectivity, so routing must be dynamic and differs from the fixed‑route assumption of MANETs. We propose a History Based Prediction Routing (HBPR) protocol for infrastructure‑less OppNets that uses node behavioral history to select the best next hop. The protocol was evaluated against Epidemic routing in simulation studies. Simulations show HBPR delivers more messages and has a lower overhead ratio than Epidemic routing.

Abstract

In Opportunistic Networks (OppNets), the existence of an end-to-end connected path between the sender and the receiver is not possible. Thus routing in this type of networks is different from the traditional Mobile Adhoc Networks (MANETs). MANETs assume the existence of a fixed route between the sender and the receiver before the start of the communication and till its completion. Routes are constructed dynamically as the source node or an intermediate node can choose any node as next hop from a group of neighbors assuming that it will take the message closer to the destination node or deliver to the destination itself. In this paper, we proposed a novel History Based Prediction Routing (HBPR) protocol for infrastructure-less OppNets which utilizes the behavioral information of the nodes to find the best next node for routing. The proposed protocol was compared with the Epidemic routing protocol. Through simulations it was found that the HBPR performs better in terms of number of messages delivered and the overhead ratio than the Epidemic protocol.