Source Capture Time Analysis of Privacy Communication Protocols for Wireless Sensor Networks

Communication privacy techniques that protect the locations of source sensor nodes and sink nodes from either global or local adversaries have received significant attention recently. The improvement in capture time, which is the time it takes for an adversary to identify the location of the source, is often estimated using simulations. In this paper, we present probabilistic models to analyze the expected capture time of source sensor nodes for recently privacy protocols against (1) local adversary using phantom routing, and (2) global adversary using statistically strong source anonymity (SSA). Using these models, we show that both phantom routing and SSA fall short of achieving high degree of anonymity. The phantom source routing improves the capture time over normal routing methods, but falls short of possible upper bounds on expected capture time. SSA is prone to simple time correlation attack by a global adversary; even with a large number of dummy message transmitted by non-source nodes, the propagation of one real message from source to sink will be sufficient to identify the source with almost no false positives. Based on these findings, we suggest a simple modification to SSA that increases the false positive rate to nearly 90 percent and thus the capture time.