Evaluation of Queueing Policies and Forwarding Strategies for Routing in Intermittently Connected Networks

Delay tolerant networking (DTN), and more specifically the subset known as intermittently connected networking, is emerging as a solution for supporting asynchronous data transfers in challenging environments where a fully connected end-to-end path between a source and destination may never exist. Message delivery in such networks is enabled via scheduled or opportunistic communication based on transitive local connectivity among nodes influenced by factors such as node mobility. Given the inherently store-and-forward and opportunistic nature of the DTN architecture, the choice of buffer management policies and message forwarding strategies can have a major impact on system performance. In this paper, we propose and evaluate different combinations of queueing policies and forwarding strategies for intermittently connected networks. We show that a probabilistic routing approach along with the correct choice of buffer management policy and forwarding strategy can result in much performance improvements in terms of message delivery, overhead and end-to-end delay