Fault Tolerance by Quartile Method with Outlier Detection in Wireless Sensor and Actor Networks

Abstract

In the emerging area of WSAN, sensors gather
information about the real world and actor performs appropriate actions. Sensors are prone to failure due to energy depletion,
hardware failure and so on. For the enhancement of fault tolerance we integrate a new heuristic approach outlier detection method. To
identify the outlier’s clusters based on sensor’s current reading and historical reading’s within a past period of time. Nowadays most research focus on the communication link fault tolerance without considering sensing fault tolerance. If the sensing data is error, then
the actor may perform incorrect action. To solve this issue Fault tolerance by quartile method (FTQM) is proposed in this paper. In FTQM, it not only determines the correct data range but also sifts the correct sensors by data discreteness. Therefore actor could perform the appropriate actions. It also leads to the reasonable communication overhead and power consumption.