Statistical Test for Distributed Cluster-based Intermittent Fault Detection in Wireless Sensor Networks

Due to the regular growing number of applications, distributed WSN enhanced to enjoy very large appeal. Occasionally, a small number of distributed nodes are drop some data value or unable to transmit accurate data to the respective or specific fusion Centre or to a nearby node. In the process of distributed network's performance suffers significantly because it is not aware of the node's malfunction. Data from the defective or affected node might be treated as outliers. So that for identify the data sent by the problematic sensor node can be found using a different statistical test. In this paper, proposed a DSFDOA method that a clustering-based statistical Z-test and a self-detectable distributed fault detection system are bringing forward. Put forward in DSFDOA and distributed sensor node algorithm is perform in Python and MATLAB, and staging is measured in terms of different parameter like false data alarm rate and detection data accuracy (DDA) and (FDAR). The results initiate demonstrate that the proposed approach shown performs better than the current algorithms and when compared to the execute results of the previous algorithm.