Gd-car: A genetic algorithm based dynamic context aware routing protocol for opportunistic networks

Abstract

Opportunistic Networks (OppNets) are an extension of Mobile Ad Hoc Networks (MANETs) in which high mobility of nodes, unawareness of network topology, and unstable node to node connections are very common. These factors inhibit the implementation of traditional routing protocols in OppNets. Hence, an efficient routing technique for OppNets is required to make communication faster, efficient and reliable. In this paper, an advanced context based routing protocol named as GD-CAR is proposed. The GD-CAR protocol uses the genetic search algorithm to predict the path for a message by dynamically updating the context information. This information is stored by each node regarding the environment in which it is operating. A fitness criterion is used to parameterise the predicted path which is finally used to evaluate the efficiency of neighbouring nodes. Simulation results show that GD-CAR protocol outperforms Prophet and GAER protocols in terms of messages delivered and average buffer time while the results for overhead ratio are comparable.