CORA: Correlation-based Resilient Aggregation in Sensor Networks

In this paper we consider the problem of resilient data aggregation, namely, when aggregation has to be performed oil a compromised sample. We present it statistical framework that is designed to mitigate the effects of an attacker who is able to alter the values of the measured parameters of the environment around some of the sensor nodes. Our proposed framework takes advantage of the naturally existing correlation between the sample elements, which is very rarely considered in other sensor network related papers. The algorithms presented are to be applied without assumption on the sensor network's sampling distribution or oil the behaviour of the attacker. The effectiveness of the algorithms is formally evaluated.