E3PR: Efficient Privacy Preserving Prediction-based Routing in Mobile Delay Tolerant Networks

Message routing is one of the major challenges in Mobile Delay Tolerant Networks (MDTNs) due to frequent and long-term network partitions. A large number of routing protocols for MDTNs belong to the category of predictionbased routing protocols, which utilize the encounter probability of nodes to guide message forwarding. However, these prediction-based routing protocols compromise the privacy of the nodes by revealing their mobility patterns. In this paper, we propose a privacy preserving prediction-based routing protocol that forwards messages by comparing aggregated information about communities instead of individual nodes. Specifically, it compares the probability that at least one node in a community will encounter the destination node. We present theoretical security analyses as well as practical performance evaluations. Our simulations on a well established community-based mobility model demonstrate that our routing protocol has comparable performance to existing prediction-based protocols. Additionally, the community information is computed efficiently and independently of the routing protocol.