Diagnosis of Wireless Sensor Networks in Presence of Permanent and Intermittent Faults

摘要: In this paper we consider the problem of distributed fault diagnosis in Wireless Sensor Networks (WSNs). The proposed Fault Diagnosis Algorithm (FDA) aims to handle both permanent and intermittent faults. sensor nodes with communication faults can be diagnosed by using conventional time-out mechanism. contrast, it is difficult detect due their inherent unpredictable behavior. FDA based on comparison measurements residual energy values neighboring nodes, exploiting spatial correlations. To faults, comparisons are made for $$r$$ r rounds. Two special cases considered: one, when an intermittently faulty node sends similar measurement value some its neighbors all rounds; another, these values, either or which deviates significantly from that Through extensive simulation analysis, scheme proved correct, complete, efficient hence, well suited WSNs.