From Contacts to Graphs: Pitfalls in Using Complex Network Analysis for DTN Routing

Delay Tolerant Networks (DTN) are networks of self-organizing wireless nodes, where end-to-end connectivity is intermittent. In these networks, forwarding decisions are made using locally collected knowledge about node behavior (e.g., past contacts between nodes) to predict which nodes are likely to deliver a content or bring it closer to the destination. One promising way of predicting future contact opportunities is to aggregate contacts seen in the past to a social graph and use metrics from complex network analysis (e.g., centrality and similarity) to assess the utility of a node to carry a piece of content. This aggregation presents an inherent tradeoff between the amount of time-related information lost during this mapping and the predictive capability of complex network analysis in this context. In this paper, we use two recent DTN routing algorithms that rely on such complex network analysis, to show that contact aggregation significantly affects the performance of these protocols. We then propose simple contact mapping algorithms that demonstrate improved performance up to a factor of 4 in delivery ratio, and robustness to various connectivity scenarios for both protocols.