Probabilistic Routing in Intermittently Connected Networks

Abstract

We consider the problem of routing in intermittently connected networks. In such networks there is no guarantee that a fully connected path between source and destination exist at any time, rendering traditional routing protocols unable to deliver messages between hosts. We propose a probabilistic routing protocol for such networks.

Routing in a cyclic mobispace

“…In [2,3], optimal delivery paths in a DTN can be discovered by constructing a time-space graph with oracles. In practical situations where no oracle is available to reveal future contacts, opportunistic routing [4,5] is proposed in which one or more copies of a message is sent along different paths and each copy is always forwarded to the node that has a higher delivery probability. Metrics for delivery probability can be either short-term metrics (e.g., the time that has elapsed since the last encounter), which have short life-spans and require frequent updates, or long-term metrics which are relatively stable over time.…”

“…Previous works have proposed a variety of long-term metrics including encounter frequency [5] and social similarity [6]. One advantage of the long-term delivery probability metrics is that they are relatively stable once generated from historical connectivity information or prior knowledge on the contact pattern of the nodes, avoiding expensive and frequent updates.…”

A Survey: Crucial Crime Information Sharing Mechanism Using DTN for Rural Areas

“…They showed in their simulations that if the buffer size is small, the packet delivery ratio of CAR is better than that of epidemic routing due to that CAR only creates a single copy for each message. Burgess, et al, [10] proposed a protocol called MaxProp for effective routing of messages. A node uses MaxProp to schedule packets transmission to its peers and determines which packets should be deleted when buffer space is almost full.…”

“…Whenever two nodes come into communication range, they exchange summary vectors that contain a compact unambiguous representation of the messages currently stored in the local buffers [9,10]. The dissemination process is somehow bounded because each message is assigned a hop count limit giving the maximum number of hops it is allowed to traverse till the destination.…”

“…Instead, nodes estimate the probability of each link to destination Lindgren, et al, proposed a probabilistic routing protocol, PRoPHET (Probabilistic Routing Protocol using History of Encounters and Transitivity) Routing: PROPHET estimates a probabilistic metric called delivery predictability [10]. It is an evolution of epidemic that introduces the concept of delievery predictability.…”