Quantifying eavesdropping vulnerability in sensor networks Madhukar Anand, Zachary Ives, and Insup Lee   View Paper (PDF)   Return to Security Abstract With respect to security, sensor networks have a number of considerations that separate them from traditional distributed systems. First, sensor devices are typically vulnerable to physical compromise. Second, they have significant power and processing constraints. Third, the most critical security issue is protecting the (statistically derived) aggregate output of the system, even if individual nodes may be compromised. We suggest that these considerations merit a rethinking of traditional security techniques: rather than depending on the resilience of cryptographic techniques, in this paper we develop new techniques to tolerate compromised nodes and to even mislead an adversary. We present our initial work on probabilistically quantifying the security of sensor network protocols, with respect to sensor data distributions and network topologies. Beginning with a taxonomy of attacks based on an adversary's goals, we focus on how to evaluate the vulnerability of sensor network protocols to eavesdropping. Different topologies and aggregation functions provide different probabilistic guarantees about system security, and make different trade-offs in power and accuracy. ©2006 Association for Computing Machinery