Energy-aware test connection assignment for the diagnosis of a wireless sensor network

In a Wireless Sensor Network (WSN) where the sensors raise alarms in response to anomalous events, we consider the problem of detecting false alarms (i.e. false positives) by using a system-level diagnosis approach. In particular, we assume that a set of t sensors in a geographic neighborhood generate the alarm, and the sink issues to the WSN a self test task that involves a number of reciprocal tests among the sensors in the WSN. Based on the test outcomes, the sink executes the diagnosis in order to identify the faulty sensors. This work presents an algorithm for the assignment of the tests (i.e. a strategy for the assignment of reciprocal tests among sensors) that assures the desired system diagnosability and that is aware of the energy consumption of the sensors. Simulation experiments show that, as compared to the existing approaches, our approach enables consistent energy savings on the sensors.