A Bayesian Trust-Based Secured Routing Protocol for Opportunistic Networks

Opportunistic networks (OppNets) allow data communication without a fixed infrastructure, making them useful in remote or disaster-hit areas. However, their open nature and intermittent connectivity make them vulnerable to malicious attacks. In this paper, a Secured Routing Protocol based on Bayesian trust is proposed for OppNets (called the Bayesian Attack-Defense Trust Model Against Nodes (BATMAN)), which uses Bayesian game theory to model the attacker's behavior under uncertainty. In such scheme, an adaptive trust management strategy is used to identify and avoid malicious nodes during the message forwarding process among the nodes. The proposed scheme is evaluated using the MIT Reality Mining dataset and its performance is compared with that of the Prophet, MaxProp, Spray-and-Wait, and Epidemic protocols, showing its superiority in terms of delivery ratio, latency, overhead ratio, and resilience, in the presence of increasing malicious behavior.