A Bayesian Routing Framework for Delay Tolerant Networks

Routing in delay tolerant networks (DTN) can benefit from the fact that most real life DTN, especially in the context of people-centric networks (e.g. Pocket Switching Networks (PSN)), exhibit some sort of periodicity in their mobility patterns. For example, public transportation networks follow periodic schedules. Even most individuals have fairly repetitive movement patterns, for example, driving to and from work at approximately the same time everyday. This paper proposes a Bayesian classifier based DTN routing framework that adopts a methodical approach for computing the routing metrics by utilizing the network parameters (e.g. spatial and temporal information at the time of packet forwarding) that capture the periodic behavior of DTN nodes. After the calculation of routing metrics, different routing instantiations are possible based on this framework. We simulate a real-world vehicular DTN network using mobility traces from a metropolitan public transportation bus network and demonstrate that even a simplistic single-copy forwarding scheme based on our framework outperforms existing gradient-based single copy schemes by 25% in terms of delivery ratio. To the best of our knowledge this work is one of the first studies that adopts Bayesian inference in the context of DTN routing.